PaliGemma/[PaliGemma_2]Convert_PaliGemma2_to_ONNX.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AQOtjPKVtaTz"
      },
      "source": [
        "##### Copyright 2025 Google LLC."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "tSzX2ErItdCq"
      },
      "outputs": [],
      "source": [
        "# @title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l8xAFcwbK14R"
      },
      "source": [
        "#### Developed by AI/ML GDE [Nitin Tiwari](https://linkedin.com/in/tiwari-nitin).\n",
        "* LinkedIn: [linkedin.com/in/tiwari-nitin](https://linkedin.com/in/tiwari-nitin)\n",
        "* GitHub: [github.com/NSTiwari](https://github.com/NSTiwari)\n",
        "* X: [@NSTiwari21](https://x.com/NSTiwari21)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xFsuF312vGxT"
      },
      "source": [
        "## Convert PaliGemma 2 to ONNX and inference on the browser using Transformers.js"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rFb1NC4OvAYU"
      },
      "source": [
        "This notebook covers Part 1 of the implementation for converting and quantizing the PaliGemma 2 Vision Language Model to ONNX for inference with Transformers.js.\n",
        "\n",
        "* [Part 1]: [Convert and quantize PaliGemma 2 to ONNX.](https://github.com/google-gemini/gemma-cookbook/blob/main/PaliGemma/[PaliGemma_2]Convert_PaliGemma2_to_ONNX.ipynb)\n",
        "\n",
        "* [Part 2]: [Inference the converted model using 🤗 Transformers.js for tasks like image captioning, zero-shot object detection, OCR, and visual Q&A.](https://github.com/google-gemini/gemma-cookbook/blob/main/PaliGemma/[PaliGemma_2]Inference_PaliGemma2_with_Transformers_js.ipynb)\n",
        "\n",
        "<table align=\"left\">\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/google-gemini/gemma-cookbook/blob/main/PaliGemma/[PaliGemma_2]Convert_PaliGemma2_to_ONNX.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
        "  </td>\n",
        "</table>\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kABIsmARuS5z"
      },
      "source": [
        "### Get access to PaliGemma 2\n",
        "\n",
        "Before using PaliGemma 2 for the first time, you must request access to the model through Hugging Face by completing the following steps:\n",
        "\n",
        "1. Log in to [Hugging Face](https://huggingface.co), or create a new Hugging Face account if you don't already have one.\n",
        "2. Go to the [PaliGemma 2 model card](https://huggingface.co/google/paligemma2-3b-pt-224) to get access to the model.\n",
        "3. Complete the consent form and accept the terms and conditions.\n",
        "\n",
        "To generate a Hugging Face token, open your [**Settings** page in Hugging Face](https://huggingface.co/settings), choose **Access Tokens** option in the left pane and click **New token**. In the next window that appears, give a name to your token and choose the type as **Write** to get the write access.\n",
        "\n",
        "Then, in Colab, select **Secrets** (🔑) in the left pane and add your Hugging Face token. Store your Hugging Face token under the name `HF_TOKEN`."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bEtSy9jkulOs"
      },
      "source": [
        "### Select the runtime\n",
        "\n",
        "To complete this tutorial, you'll need to have a Colab runtime with sufficient resources to load the PaliGemma 2 model. In this case, you need at least an L4 GPU:\n",
        "\n",
        "1. In the upper-right of the Colab window, click the **▾ (Additional connection options)** dropdown menu.\n",
        "1. Select **Change runtime type**.\n",
        "1. Under **Hardware accelerator**, select **L4 GPU**."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G5_-49m8LMUJ"
      },
      "source": [
        "### Step 1: Install libraries and dependencies\n",
        "*Note: You might need to restart the runtime after the cell finishes execution.*"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WJ1gQE0-KtHh"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "  Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "  Building wheel for transformers (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting optimum[exporters]\n",
            "  Downloading optimum-1.24.0-py3-none-any.whl.metadata (21 kB)\n",
            "Requirement already satisfied: transformers>=4.29 in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (4.50.0.dev0)\n",
            "Requirement already satisfied: torch>=1.11 in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (2.5.1+cu124)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (24.2)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (1.26.4)\n",
            "Requirement already satisfied: huggingface-hub>=0.8.0 in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (0.28.1)\n",
            "Collecting onnx (from optimum[exporters])\n",
            "  Downloading onnx-1.17.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (16 kB)\n",
            "Collecting onnxruntime (from optimum[exporters])\n",
            "  Downloading onnxruntime-1.20.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (4.5 kB)\n",
            "Requirement already satisfied: timm in /usr/local/lib/python3.11/dist-packages (from optimum[exporters]) (1.0.14)\n",
            "Collecting transformers>=4.29 (from optimum[exporters])\n",
            "  Downloading transformers-4.48.3-py3-none-any.whl.metadata (44 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m44.4/44.4 kB\u001b[0m \u001b[31m4.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (3.17.0)\n",
            "Requirement already satisfied: fsspec>=2023.5.0 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (2024.10.0)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (6.0.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (2.32.3)\n",
            "Requirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (4.67.1)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[exporters]) (4.12.2)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (3.4.2)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (3.1.5)\n",
            "Collecting nvidia-cuda-nvrtc-cu12==12.4.127 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cuda_nvrtc_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Collecting nvidia-cuda-runtime-cu12==12.4.127 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cuda_runtime_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Collecting nvidia-cuda-cupti-cu12==12.4.127 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cuda_cupti_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl.metadata (1.6 kB)\n",
            "Collecting nvidia-cudnn-cu12==9.1.0.70 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cudnn_cu12-9.1.0.70-py3-none-manylinux2014_x86_64.whl.metadata (1.6 kB)\n",
            "Collecting nvidia-cublas-cu12==12.4.5.8 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cublas_cu12-12.4.5.8-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Collecting nvidia-cufft-cu12==11.2.1.3 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cufft_cu12-11.2.1.3-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Collecting nvidia-curand-cu12==10.3.5.147 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_curand_cu12-10.3.5.147-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Collecting nvidia-cusolver-cu12==11.6.1.9 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cusolver_cu12-11.6.1.9-py3-none-manylinux2014_x86_64.whl.metadata (1.6 kB)\n",
            "Collecting nvidia-cusparse-cu12==12.3.1.170 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_cusparse_cu12-12.3.1.170-py3-none-manylinux2014_x86_64.whl.metadata (1.6 kB)\n",
            "Requirement already satisfied: nvidia-nccl-cu12==2.21.5 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (2.21.5)\n",
            "Requirement already satisfied: nvidia-nvtx-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (12.4.127)\n",
            "Collecting nvidia-nvjitlink-cu12==12.4.127 (from torch>=1.11->optimum[exporters])\n",
            "  Downloading nvidia_nvjitlink_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\n",
            "Requirement already satisfied: triton==3.1.0 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (3.1.0)\n",
            "Requirement already satisfied: sympy==1.13.1 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[exporters]) (1.13.1)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from sympy==1.13.1->torch>=1.11->optimum[exporters]) (1.3.0)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[exporters]) (2024.11.6)\n",
            "Requirement already satisfied: tokenizers<0.22,>=0.21 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[exporters]) (0.21.0)\n",
            "Requirement already satisfied: safetensors>=0.4.1 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[exporters]) (0.5.2)\n",
            "Requirement already satisfied: protobuf>=3.20.2 in /usr/local/lib/python3.11/dist-packages (from onnx->optimum[exporters]) (4.25.6)\n",
            "Collecting coloredlogs (from onnxruntime->optimum[exporters])\n",
            "  Downloading coloredlogs-15.0.1-py2.py3-none-any.whl.metadata (12 kB)\n",
            "Requirement already satisfied: flatbuffers in /usr/local/lib/python3.11/dist-packages (from onnxruntime->optimum[exporters]) (25.2.10)\n",
            "Requirement already satisfied: torchvision in /usr/local/lib/python3.11/dist-packages (from timm->optimum[exporters]) (0.20.1+cu124)\n",
            "Collecting humanfriendly>=9.1 (from coloredlogs->onnxruntime->optimum[exporters])\n",
            "  Downloading humanfriendly-10.0-py2.py3-none-any.whl.metadata (9.2 kB)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/dist-packages (from jinja2->torch>=1.11->optimum[exporters]) (3.0.2)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.8.0->optimum[exporters]) (3.4.1)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.8.0->optimum[exporters]) (3.10)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.8.0->optimum[exporters]) (2.3.0)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/dist-packages (from requests->huggingface-hub>=0.8.0->optimum[exporters]) (2025.1.31)\n",
            "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.11/dist-packages (from torchvision->timm->optimum[exporters]) (11.1.0)\n",
            "Downloading nvidia_cublas_cu12-12.4.5.8-py3-none-manylinux2014_x86_64.whl (363.4 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m363.4/363.4 MB\u001b[0m \u001b[31m2.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cuda_cupti_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl (13.8 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m13.8/13.8 MB\u001b[0m \u001b[31m104.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cuda_nvrtc_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl (24.6 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m24.6/24.6 MB\u001b[0m \u001b[31m78.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cuda_runtime_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl (883 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m883.7/883.7 kB\u001b[0m \u001b[31m64.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cudnn_cu12-9.1.0.70-py3-none-manylinux2014_x86_64.whl (664.8 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m664.8/664.8 MB\u001b[0m \u001b[31m1.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cufft_cu12-11.2.1.3-py3-none-manylinux2014_x86_64.whl (211.5 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m211.5/211.5 MB\u001b[0m \u001b[31m4.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_curand_cu12-10.3.5.147-py3-none-manylinux2014_x86_64.whl (56.3 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m56.3/56.3 MB\u001b[0m \u001b[31m36.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cusolver_cu12-11.6.1.9-py3-none-manylinux2014_x86_64.whl (127.9 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m127.9/127.9 MB\u001b[0m \u001b[31m16.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_cusparse_cu12-12.3.1.170-py3-none-manylinux2014_x86_64.whl (207.5 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m207.5/207.5 MB\u001b[0m \u001b[31m4.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_nvjitlink_cu12-12.4.127-py3-none-manylinux2014_x86_64.whl (21.1 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m21.1/21.1 MB\u001b[0m \u001b[31m93.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading transformers-4.48.3-py3-none-any.whl (9.7 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9.7/9.7 MB\u001b[0m \u001b[31m121.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading onnx-1.17.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (16.0 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m16.0/16.0 MB\u001b[0m \u001b[31m107.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading onnxruntime-1.20.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (13.3 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m13.3/13.3 MB\u001b[0m \u001b[31m112.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading optimum-1.24.0-py3-none-any.whl (433 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m433.6/433.6 kB\u001b[0m \u001b[31m35.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading coloredlogs-15.0.1-py2.py3-none-any.whl (46 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m46.0/46.0 kB\u001b[0m \u001b[31m4.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading humanfriendly-10.0-py2.py3-none-any.whl (86 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m86.8/86.8 kB\u001b[0m \u001b[31m8.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: onnx, nvidia-nvjitlink-cu12, nvidia-curand-cu12, nvidia-cufft-cu12, nvidia-cuda-runtime-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-cupti-cu12, nvidia-cublas-cu12, humanfriendly, nvidia-cusparse-cu12, nvidia-cudnn-cu12, coloredlogs, onnxruntime, nvidia-cusolver-cu12, transformers, optimum\n",
            "  Attempting uninstall: nvidia-nvjitlink-cu12\n",
            "    Found existing installation: nvidia-nvjitlink-cu12 12.5.82\n",
            "    Uninstalling nvidia-nvjitlink-cu12-12.5.82:\n",
            "      Successfully uninstalled nvidia-nvjitlink-cu12-12.5.82\n",
            "  Attempting uninstall: nvidia-curand-cu12\n",
            "    Found existing installation: nvidia-curand-cu12 10.3.6.82\n",
            "    Uninstalling nvidia-curand-cu12-10.3.6.82:\n",
            "      Successfully uninstalled nvidia-curand-cu12-10.3.6.82\n",
            "  Attempting uninstall: nvidia-cufft-cu12\n",
            "    Found existing installation: nvidia-cufft-cu12 11.2.3.61\n",
            "    Uninstalling nvidia-cufft-cu12-11.2.3.61:\n",
            "      Successfully uninstalled nvidia-cufft-cu12-11.2.3.61\n",
            "  Attempting uninstall: nvidia-cuda-runtime-cu12\n",
            "    Found existing installation: nvidia-cuda-runtime-cu12 12.5.82\n",
            "    Uninstalling nvidia-cuda-runtime-cu12-12.5.82:\n",
            "      Successfully uninstalled nvidia-cuda-runtime-cu12-12.5.82\n",
            "  Attempting uninstall: nvidia-cuda-nvrtc-cu12\n",
            "    Found existing installation: nvidia-cuda-nvrtc-cu12 12.5.82\n",
            "    Uninstalling nvidia-cuda-nvrtc-cu12-12.5.82:\n",
            "      Successfully uninstalled nvidia-cuda-nvrtc-cu12-12.5.82\n",
            "  Attempting uninstall: nvidia-cuda-cupti-cu12\n",
            "    Found existing installation: nvidia-cuda-cupti-cu12 12.5.82\n",
            "    Uninstalling nvidia-cuda-cupti-cu12-12.5.82:\n",
            "      Successfully uninstalled nvidia-cuda-cupti-cu12-12.5.82\n",
            "  Attempting uninstall: nvidia-cublas-cu12\n",
            "    Found existing installation: nvidia-cublas-cu12 12.5.3.2\n",
            "    Uninstalling nvidia-cublas-cu12-12.5.3.2:\n",
            "      Successfully uninstalled nvidia-cublas-cu12-12.5.3.2\n",
            "  Attempting uninstall: nvidia-cusparse-cu12\n",
            "    Found existing installation: nvidia-cusparse-cu12 12.5.1.3\n",
            "    Uninstalling nvidia-cusparse-cu12-12.5.1.3:\n",
            "      Successfully uninstalled nvidia-cusparse-cu12-12.5.1.3\n",
            "  Attempting uninstall: nvidia-cudnn-cu12\n",
            "    Found existing installation: nvidia-cudnn-cu12 9.3.0.75\n",
            "    Uninstalling nvidia-cudnn-cu12-9.3.0.75:\n",
            "      Successfully uninstalled nvidia-cudnn-cu12-9.3.0.75\n",
            "  Attempting uninstall: nvidia-cusolver-cu12\n",
            "    Found existing installation: nvidia-cusolver-cu12 11.6.3.83\n",
            "    Uninstalling nvidia-cusolver-cu12-11.6.3.83:\n",
            "      Successfully uninstalled nvidia-cusolver-cu12-11.6.3.83\n",
            "  Attempting uninstall: transformers\n",
            "    Found existing installation: transformers 4.50.0.dev0\n",
            "    Uninstalling transformers-4.50.0.dev0:\n",
            "      Successfully uninstalled transformers-4.50.0.dev0\n",
            "Successfully installed coloredlogs-15.0.1 humanfriendly-10.0 nvidia-cublas-cu12-12.4.5.8 nvidia-cuda-cupti-cu12-12.4.127 nvidia-cuda-nvrtc-cu12-12.4.127 nvidia-cuda-runtime-cu12-12.4.127 nvidia-cudnn-cu12-9.1.0.70 nvidia-cufft-cu12-11.2.1.3 nvidia-curand-cu12-10.3.5.147 nvidia-cusolver-cu12-11.6.1.9 nvidia-cusparse-cu12-12.3.1.170 nvidia-nvjitlink-cu12-12.4.127 onnx-1.17.0 onnxruntime-1.20.1 optimum-1.24.0 transformers-4.48.3\n",
            "Collecting onnxslim\n",
            "  Downloading onnxslim-0.1.48-py3-none-any.whl.metadata (4.6 kB)\n",
            "Requirement already satisfied: onnx in /usr/local/lib/python3.11/dist-packages (from onnxslim) (1.17.0)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.11/dist-packages (from onnxslim) (1.13.1)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxslim) (24.2)\n",
            "Requirement already satisfied: numpy>=1.20 in /usr/local/lib/python3.11/dist-packages (from onnx->onnxslim) (1.26.4)\n",
            "Requirement already satisfied: protobuf>=3.20.2 in /usr/local/lib/python3.11/dist-packages (from onnx->onnxslim) (4.25.6)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from sympy->onnxslim) (1.3.0)\n",
            "Downloading onnxslim-0.1.48-py3-none-any.whl (142 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m142.9/142.9 kB\u001b[0m \u001b[31m12.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: onnxslim\n",
            "Successfully installed onnxslim-0.1.48\n",
            "Collecting onnxconverter_common\n",
            "  Downloading onnxconverter_common-1.14.0-py2.py3-none-any.whl.metadata (4.2 kB)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from onnxconverter_common) (1.26.4)\n",
            "Requirement already satisfied: onnx in /usr/local/lib/python3.11/dist-packages (from onnxconverter_common) (1.17.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxconverter_common) (24.2)\n",
            "Collecting protobuf==3.20.2 (from onnxconverter_common)\n",
            "  Downloading protobuf-3.20.2-py2.py3-none-any.whl.metadata (720 bytes)\n",
            "Downloading onnxconverter_common-1.14.0-py2.py3-none-any.whl (84 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m84.5/84.5 kB\u001b[0m \u001b[31m9.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading protobuf-3.20.2-py2.py3-none-any.whl (162 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m162.1/162.1 kB\u001b[0m \u001b[31m18.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: protobuf, onnxconverter_common\n",
            "  Attempting uninstall: protobuf\n",
            "    Found existing installation: protobuf 4.25.6\n",
            "    Uninstalling protobuf-4.25.6:\n",
            "      Successfully uninstalled protobuf-4.25.6\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "tensorflow-metadata 1.16.1 requires protobuf<6.0.0dev,>=4.25.2; python_version >= \"3.11\", but you have protobuf 3.20.2 which is incompatible.\n",
            "tensorflow 2.18.0 requires protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<6.0.0dev,>=3.20.3, but you have protobuf 3.20.2 which is incompatible.\n",
            "grpcio-status 1.62.3 requires protobuf>=4.21.6, but you have protobuf 3.20.2 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0mSuccessfully installed onnxconverter_common-1.14.0 protobuf-3.20.2\n"
          ]
        },
        {
          "data": {
            "application/vnd.colab-display-data+json": {
              "id": "25e17eaf03e94ce5a5efa763ff7b1f2e",
              "pip_warning": {
                "packages": [
                  "google"
                ]
              }
            }
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting onnx_graphsurgeon==0.5.2\n",
            "  Downloading onnx_graphsurgeon-0.5.2-py2.py3-none-any.whl.metadata (8.1 kB)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from onnx_graphsurgeon==0.5.2) (1.26.4)\n",
            "Requirement already satisfied: onnx>=1.14.0 in /usr/local/lib/python3.11/dist-packages (from onnx_graphsurgeon==0.5.2) (1.17.0)\n",
            "Requirement already satisfied: protobuf>=3.20.2 in /usr/local/lib/python3.11/dist-packages (from onnx>=1.14.0->onnx_graphsurgeon==0.5.2) (3.20.2)\n",
            "Downloading onnx_graphsurgeon-0.5.2-py2.py3-none-any.whl (56 kB)\n",
            "\u001b[?25l   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/56.4 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m56.4/56.4 kB\u001b[0m \u001b[31m6.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: onnx_graphsurgeon\n",
            "Successfully installed onnx_graphsurgeon-0.5.2\n",
            "Requirement already satisfied: onnxruntime in /usr/local/lib/python3.11/dist-packages (1.20.1)\n",
            "Requirement already satisfied: coloredlogs in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (15.0.1)\n",
            "Requirement already satisfied: flatbuffers in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (25.2.10)\n",
            "Requirement already satisfied: numpy>=1.21.6 in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (1.26.4)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (24.2)\n",
            "Requirement already satisfied: protobuf in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (3.20.2)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (1.13.1)\n",
            "Requirement already satisfied: humanfriendly>=9.1 in /usr/local/lib/python3.11/dist-packages (from coloredlogs->onnxruntime) (10.0)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from sympy->onnxruntime) (1.3.0)\n",
            "Collecting onnxruntime-tools\n",
            "  Downloading onnxruntime_tools-1.7.0-py3-none-any.whl.metadata (14 kB)\n",
            "Requirement already satisfied: onnx in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (1.17.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (1.26.4)\n",
            "Requirement already satisfied: coloredlogs in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (15.0.1)\n",
            "Requirement already satisfied: psutil in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (5.9.5)\n",
            "Requirement already satisfied: py-cpuinfo in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (9.0.0)\n",
            "Collecting py3nvml (from onnxruntime-tools)\n",
            "  Downloading py3nvml-0.2.7-py3-none-any.whl.metadata (13 kB)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxruntime-tools) (24.2)\n",
            "Requirement already satisfied: humanfriendly>=9.1 in /usr/local/lib/python3.11/dist-packages (from coloredlogs->onnxruntime-tools) (10.0)\n",
            "Requirement already satisfied: protobuf>=3.20.2 in /usr/local/lib/python3.11/dist-packages (from onnx->onnxruntime-tools) (3.20.2)\n",
            "Collecting xmltodict (from py3nvml->onnxruntime-tools)\n",
            "  Downloading xmltodict-0.14.2-py2.py3-none-any.whl.metadata (8.0 kB)\n",
            "Downloading onnxruntime_tools-1.7.0-py3-none-any.whl (212 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m212.7/212.7 kB\u001b[0m \u001b[31m21.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading py3nvml-0.2.7-py3-none-any.whl (55 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m55.5/55.5 kB\u001b[0m \u001b[31m6.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading xmltodict-0.14.2-py2.py3-none-any.whl (10.0 kB)\n",
            "Installing collected packages: xmltodict, py3nvml, onnxruntime-tools\n",
            "Successfully installed onnxruntime-tools-1.7.0 py3nvml-0.2.7 xmltodict-0.14.2\n",
            "Requirement already satisfied: optimum[onnxruntime] in /usr/local/lib/python3.11/dist-packages (1.24.0)\n",
            "Requirement already satisfied: transformers>=4.29 in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (4.48.3)\n",
            "Requirement already satisfied: torch>=1.11 in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (2.5.1+cu124)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (24.2)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (1.26.4)\n",
            "Requirement already satisfied: huggingface-hub>=0.8.0 in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (0.28.1)\n",
            "Requirement already satisfied: onnx in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (1.17.0)\n",
            "Requirement already satisfied: onnxruntime>=1.11.0 in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (1.20.1)\n",
            "Collecting datasets>=1.2.1 (from optimum[onnxruntime])\n",
            "  Downloading datasets-3.3.1-py3-none-any.whl.metadata (19 kB)\n",
            "Collecting evaluate (from optimum[onnxruntime])\n",
            "  Downloading evaluate-0.4.3-py3-none-any.whl.metadata (9.2 kB)\n",
            "Requirement already satisfied: protobuf>=3.20.1 in /usr/local/lib/python3.11/dist-packages (from optimum[onnxruntime]) (3.20.2)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (3.17.0)\n",
            "Requirement already satisfied: pyarrow>=15.0.0 in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (17.0.0)\n",
            "Collecting dill<0.3.9,>=0.3.0 (from datasets>=1.2.1->optimum[onnxruntime])\n",
            "  Downloading dill-0.3.8-py3-none-any.whl.metadata (10 kB)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (2.2.2)\n",
            "Requirement already satisfied: requests>=2.32.2 in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (2.32.3)\n",
            "Requirement already satisfied: tqdm>=4.66.3 in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (4.67.1)\n",
            "Collecting xxhash (from datasets>=1.2.1->optimum[onnxruntime])\n",
            "  Downloading xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB)\n",
            "Collecting multiprocess<0.70.17 (from datasets>=1.2.1->optimum[onnxruntime])\n",
            "  Downloading multiprocess-0.70.16-py311-none-any.whl.metadata (7.2 kB)\n",
            "Requirement already satisfied: fsspec<=2024.12.0,>=2023.1.0 in /usr/local/lib/python3.11/dist-packages (from fsspec[http]<=2024.12.0,>=2023.1.0->datasets>=1.2.1->optimum[onnxruntime]) (2024.10.0)\n",
            "Requirement already satisfied: aiohttp in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (3.11.12)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.11/dist-packages (from datasets>=1.2.1->optimum[onnxruntime]) (6.0.2)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.11/dist-packages (from huggingface-hub>=0.8.0->optimum[onnxruntime]) (4.12.2)\n",
            "Requirement already satisfied: coloredlogs in /usr/local/lib/python3.11/dist-packages (from onnxruntime>=1.11.0->optimum[onnxruntime]) (15.0.1)\n",
            "Requirement already satisfied: flatbuffers in /usr/local/lib/python3.11/dist-packages (from onnxruntime>=1.11.0->optimum[onnxruntime]) (25.2.10)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.11/dist-packages (from onnxruntime>=1.11.0->optimum[onnxruntime]) (1.13.1)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (3.4.2)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (3.1.5)\n",
            "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.127)\n",
            "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.127)\n",
            "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.127)\n",
            "Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (9.1.0.70)\n",
            "Requirement already satisfied: nvidia-cublas-cu12==12.4.5.8 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.5.8)\n",
            "Requirement already satisfied: nvidia-cufft-cu12==11.2.1.3 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (11.2.1.3)\n",
            "Requirement already satisfied: nvidia-curand-cu12==10.3.5.147 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (10.3.5.147)\n",
            "Requirement already satisfied: nvidia-cusolver-cu12==11.6.1.9 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (11.6.1.9)\n",
            "Requirement already satisfied: nvidia-cusparse-cu12==12.3.1.170 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.3.1.170)\n",
            "Requirement already satisfied: nvidia-nccl-cu12==2.21.5 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (2.21.5)\n",
            "Requirement already satisfied: nvidia-nvtx-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.127)\n",
            "Requirement already satisfied: nvidia-nvjitlink-cu12==12.4.127 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (12.4.127)\n",
            "Requirement already satisfied: triton==3.1.0 in /usr/local/lib/python3.11/dist-packages (from torch>=1.11->optimum[onnxruntime]) (3.1.0)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.11/dist-packages (from sympy->onnxruntime>=1.11.0->optimum[onnxruntime]) (1.3.0)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[onnxruntime]) (2024.11.6)\n",
            "Requirement already satisfied: tokenizers<0.22,>=0.21 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[onnxruntime]) (0.21.0)\n",
            "Requirement already satisfied: safetensors>=0.4.1 in /usr/local/lib/python3.11/dist-packages (from transformers>=4.29->optimum[onnxruntime]) (0.5.2)\n",
            "Requirement already satisfied: aiohappyeyeballs>=2.3.0 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (2.4.6)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (1.3.2)\n",
            "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (25.1.0)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (1.5.0)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (6.1.0)\n",
            "Requirement already satisfied: propcache>=0.2.0 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (0.2.1)\n",
            "Requirement already satisfied: yarl<2.0,>=1.17.0 in /usr/local/lib/python3.11/dist-packages (from aiohttp->datasets>=1.2.1->optimum[onnxruntime]) (1.18.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests>=2.32.2->datasets>=1.2.1->optimum[onnxruntime]) (3.4.1)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/dist-packages (from requests>=2.32.2->datasets>=1.2.1->optimum[onnxruntime]) (3.10)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests>=2.32.2->datasets>=1.2.1->optimum[onnxruntime]) (2.3.0)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/dist-packages (from requests>=2.32.2->datasets>=1.2.1->optimum[onnxruntime]) (2025.1.31)\n",
            "Requirement already satisfied: humanfriendly>=9.1 in /usr/local/lib/python3.11/dist-packages (from coloredlogs->onnxruntime>=1.11.0->optimum[onnxruntime]) (10.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/dist-packages (from jinja2->torch>=1.11->optimum[onnxruntime]) (3.0.2)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.11/dist-packages (from pandas->datasets>=1.2.1->optimum[onnxruntime]) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.11/dist-packages (from pandas->datasets>=1.2.1->optimum[onnxruntime]) (2025.1)\n",
            "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.11/dist-packages (from pandas->datasets>=1.2.1->optimum[onnxruntime]) (2025.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.11/dist-packages (from python-dateutil>=2.8.2->pandas->datasets>=1.2.1->optimum[onnxruntime]) (1.17.0)\n",
            "Downloading datasets-3.3.1-py3-none-any.whl (484 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m484.9/484.9 kB\u001b[0m \u001b[31m40.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading evaluate-0.4.3-py3-none-any.whl (84 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m84.0/84.0 kB\u001b[0m \u001b[31m9.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading dill-0.3.8-py3-none-any.whl (116 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m116.3/116.3 kB\u001b[0m \u001b[31m12.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading multiprocess-0.70.16-py311-none-any.whl (143 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m143.5/143.5 kB\u001b[0m \u001b[31m16.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (194 kB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.8/194.8 kB\u001b[0m \u001b[31m18.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: xxhash, dill, multiprocess, datasets, evaluate\n",
            "Successfully installed datasets-3.3.1 dill-0.3.8 evaluate-0.4.3 multiprocess-0.70.16 xxhash-3.5.0\n",
            "Requirement already satisfied: onnxconverter-common in /usr/local/lib/python3.11/dist-packages (1.14.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from onnxconverter-common) (1.26.4)\n",
            "Requirement already satisfied: onnx in /usr/local/lib/python3.11/dist-packages (from onnxconverter-common) (1.17.0)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxconverter-common) (24.2)\n",
            "Requirement already satisfied: protobuf==3.20.2 in /usr/local/lib/python3.11/dist-packages (from onnxconverter-common) (3.20.2)\n"
          ]
        }
      ],
      "source": [
        "!pip install -q --upgrade git+https://github.com/huggingface/transformers.git\n",
        "!pip install optimum[exporters]\n",
        "!pip install onnxslim\n",
        "!pip install onnxconverter_common\n",
        "!pip install onnx_graphsurgeon==0.5.2\n",
        "!pip install onnxruntime\n",
        "!pip install onnxruntime-tools\n",
        "!pip install optimum[onnxruntime]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TJ0IjMFfL8Nh"
      },
      "source": [
        "### Step 2: Setup environment\n",
        "Before we begin with the conversion of PaliGemma 2 to ONNX (Open Neural Network Exchange), we first need to include the following line of code:\n",
        "\n",
        "`GLOBAL.onnx_shape_inference = False`\n",
        "\n",
        "This should be added before line 662 in the file /usr/local/lib/python3.11/dist-packages/torch/onnx/utils.py as follows:\n",
        "\n",
        "```\n",
        "# Add the below line.\n",
        "GLOBAL.onnx_shape_inference = False\n",
        "if GLOBALS.onnx_shape_inference:\n",
        "        _C._jit_pass_onnx_graph_shape_type_inference(\n",
        "            graph, params_dict, GLOBALS.export_onnx_opset_version\n",
        "        )\n",
        "```\n",
        "\n",
        "\n",
        "\n",
        "\n",
        " This adjustment serves as a temporary workaround for a [bug](https://github.com/pytorch/pytorch/issues/147259) in PyTorch until a permanent fix is implemented.\n",
        "\n",
        "*Note: Restart the runtime for the changes to take effect.*"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ozNBzNCgPfTR"
      },
      "source": [
        "### Step 3: Convert PaliGemma 2 to ONNX\n",
        "Now, we're ready to begin the conversion process. This process involves converting the PaliGemma 2 model weights, which include:\n",
        "\n",
        "* Language Decoder (Gemma 2)\n",
        "* Vision Encoder (SigLIP)\n",
        "* Embedding Tokens"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "oPIKg0LiLj2I"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "from google.colab import userdata\n",
        "\n",
        "os.environ[\"HF_TOKEN\"] = userdata.get('HF_TOKEN')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "MHqaEySkNgjR"
      },
      "outputs": [],
      "source": [
        "# Choose the PaliGemma 2 variant.\n",
        "\n",
        "model_id = \"paligemma2-3b-mix-224\" # @param [\"paligemma2-3b-mix-224\", \"paligemma2-3b-mix-448\", \"paligemma2-3b-pt-224\", \"paligemma2-3b-ft-docci-448\", \"paligemma2-3b-pt-448\", \"paligemma2-3b-pt-896\"]\n",
        "model_id = f\"google/{model_id}\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lUw_EXnFMFcP"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Converting google/paligemma2-3b-mix-224 to ONNX.\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "fe8145a646e54b1bb31976f3d7d49992",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "config.json:   0%|          | 0.00/1.33k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "6264b3cbb47740ff97cee85199c3b353",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model.safetensors.index.json:   0%|          | 0.00/75.1k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e4ff60b6926a4bf3b29c15689f1595ac",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a7ae6ef702f74665b7f77ab449283cb9",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model-00001-of-00002.safetensors:   0%|          | 0.00/4.99G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "16665bf8c9af405aa6eec920bc61b704",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model-00002-of-00002.safetensors:   0%|          | 0.00/1.07G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "da2ad2aef86545c28d1c8dad9d483837",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "689d852b9fd14b23bb55d9dedac5e19c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "generation_config.json:   0%|          | 0.00/173 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "9c4b53a2d725493184fccbfe699ccef9",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "preprocessor_config.json:   0%|          | 0.00/424 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e4eda4c010d949fe8da283f0d01d42cf",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer_config.json:   0%|          | 0.00/243k [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "86ff7815db774a32998d1628f96a2abd",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer.json:   0%|          | 0.00/34.6M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "41d143a8d30041509d76504ba0f5582b",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "special_tokens_map.json:   0%|          | 0.00/733 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.11/dist-packages/transformers/models/gemma2/modeling_gemma2.py:625: TracerWarning: Converting a tensor to a Python number might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!\n",
            "  attention_mask.shape[-1] if attention_mask.dim() == 2 else cache_position[-1].item()\n",
            "/usr/local/lib/python3.11/dist-packages/transformers/models/gemma2/modeling_gemma2.py:640: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.\n",
            "  normalizer = torch.tensor(self.config.hidden_size**0.5, dtype=hidden_states.dtype)\n",
            "/usr/local/lib/python3.11/dist-packages/transformers/models/gemma2/modeling_gemma2.py:294: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs!\n",
            "  effective_seq_len = max(cache_position.shape[0], self.sliding_window)\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from transformers import (\n",
        "    AutoProcessor,\n",
        "    PaliGemmaForConditionalGeneration,\n",
        "    DynamicCache,\n",
        ")\n",
        "\n",
        "print(f\"Converting {model_id} to ONNX.\")\n",
        "\n",
        "def new_len(self: torch.Tensor):\n",
        "    return self.shape[0]\n",
        "\n",
        "torch.Tensor.__len__ = new_len\n",
        "\n",
        "\n",
        "class VisionEncoder(nn.Module):\n",
        "  def __init__(self, paligemma_model):\n",
        "    super().__init__()\n",
        "    self.config = paligemma_model.config\n",
        "    self.vision_tower = paligemma_model.vision_tower\n",
        "    self.multi_modal_projector = paligemma_model.multi_modal_projector\n",
        "\n",
        "  def forward(self, pixel_values: torch.FloatTensor):\n",
        "      \"\"\"\n",
        "      Obtains image last hidden states from the vision tower and apply multimodal projection.\n",
        "\n",
        "      Args:\n",
        "          pixel_values (`torch.FloatTensor]` of shape `(batch_size, channels, height, width)`)\n",
        "              The tensors corresponding to the input images.\n",
        "      Returns:\n",
        "          image_features (`torch.Tensor`): Image feature tensor of shape `(num_images, image_length, embed_dim)`).\n",
        "      \"\"\"\n",
        "      image_outputs = self.vision_tower(pixel_values)\n",
        "      selected_image_feature = image_outputs.last_hidden_state\n",
        "      image_features = self.multi_modal_projector(selected_image_feature)\n",
        "      image_features = image_features / (self.config.text_config.hidden_size**0.5)\n",
        "      return image_features\n",
        "\n",
        "\n",
        "class PatchedPaliGemmaForConditionalGeneration(PaliGemmaForConditionalGeneration):\n",
        "    def forward(self, *args):\n",
        "        inputs_embeds, position_ids, *past_key_values_args = args\n",
        "        config = model.config.text_config\n",
        "\n",
        "        # Convert past_key_values list to DynamicCache\n",
        "        if len(past_key_values_args) == 0:\n",
        "            past_key_values = None\n",
        "        else:\n",
        "            past_key_values = DynamicCache(config.num_hidden_layers)\n",
        "            for i in range(config.num_hidden_layers):\n",
        "                key = past_key_values_args.pop(0)\n",
        "                value = past_key_values_args.pop(0)\n",
        "                past_key_values.update(key_states=key, value_states=value, layer_idx=i)\n",
        "\n",
        "\n",
        "        batch_size = inputs_embeds.shape[0]\n",
        "\n",
        "        o = self.language_model.forward(\n",
        "            inputs_embeds=inputs_embeds,\n",
        "            # Create a 4D attention mask of all zeros (attend to everything)\n",
        "            attention_mask=torch.zeros(\n",
        "                batch_size,\n",
        "                1, # num_attention_heads (1 -> expand to num_attention_heads)\n",
        "                1, # sequence_length (1 -> expand to sequence_length)\n",
        "                1, # total_sequence_length (1 -> expand to total_sequence_length)\n",
        "                dtype=torch.float32,\n",
        "            ),\n",
        "            position_ids=position_ids,\n",
        "            past_key_values=past_key_values,\n",
        "        )\n",
        "\n",
        "        flattened_past_key_values_outputs = {\n",
        "            \"logits\": o.logits,\n",
        "        }\n",
        "        output_past_key_values: DynamicCache = o.past_key_values\n",
        "        for i, (key, value) in enumerate(\n",
        "            zip(output_past_key_values.key_cache, output_past_key_values.value_cache)\n",
        "        ):\n",
        "            flattened_past_key_values_outputs[f\"present.{i}.key\"] = key\n",
        "            flattened_past_key_values_outputs[f\"present.{i}.value\"] = value\n",
        "\n",
        "        return flattened_past_key_values_outputs\n",
        "\n",
        "\n",
        "# Constants\n",
        "OUTPUT_FOLDER = os.path.join(\"output\", model_id)\n",
        "TEXT_MODEL_NAME = \"decoder_model_merged.onnx\"\n",
        "VISION_MODEL_NAME = \"vision_encoder.onnx\"\n",
        "EMBED_MODEL_NAME = \"embed_tokens.onnx\"\n",
        "TEMP_MODEL_OUTPUT_FOLDER = os.path.join(OUTPUT_FOLDER, \"temp\")\n",
        "FINAL_MODEL_OUTPUT_FOLDER = os.path.join(OUTPUT_FOLDER, \"onnx\")\n",
        "\n",
        "\n",
        "# Load model and processor\n",
        "model = PatchedPaliGemmaForConditionalGeneration.from_pretrained(\n",
        "    model_id,\n",
        ").eval()\n",
        "vision_model = VisionEncoder(model)\n",
        "embed_layer = model.language_model.model.embed_tokens\n",
        "\n",
        "processor = AutoProcessor.from_pretrained(model_id)\n",
        "\n",
        "# Save model configs and processor\n",
        "model.config.save_pretrained(OUTPUT_FOLDER)\n",
        "model.generation_config.save_pretrained(OUTPUT_FOLDER)\n",
        "processor.save_pretrained(OUTPUT_FOLDER)\n",
        "os.makedirs(TEMP_MODEL_OUTPUT_FOLDER, exist_ok=True)\n",
        "\n",
        "\n",
        "# Configuration values\n",
        "## Text model\n",
        "text_config = model.config.text_config\n",
        "num_attention_heads = text_config.num_attention_heads\n",
        "num_key_value_heads = text_config.num_key_value_heads\n",
        "head_dim = text_config.head_dim\n",
        "num_layers = text_config.num_hidden_layers\n",
        "hidden_size = text_config.hidden_size\n",
        "\n",
        "# Dummy input sizes\n",
        "batch_size = 2\n",
        "sequence_length = 32\n",
        "past_sequence_length = 8\n",
        "\n",
        "## Text inputs\n",
        "dummy_past_key_values_kwargs = {\n",
        "    f\"past_key_values.{i}.{key}\": torch.zeros(\n",
        "        batch_size,\n",
        "        num_key_value_heads,\n",
        "        past_sequence_length,\n",
        "        head_dim,\n",
        "        dtype=torch.float32,\n",
        "    )\n",
        "    for i in range(num_layers)\n",
        "    for key in [\"key\", \"value\"]\n",
        "}\n",
        "inputs_embeds = torch.randn(\n",
        "    (batch_size, sequence_length, hidden_size),\n",
        ")\n",
        "\n",
        "total_sequence_length = sequence_length + past_sequence_length\n",
        "position_ids = torch.arange(1, sequence_length + 1, dtype=torch.int64).expand(batch_size, sequence_length)\n",
        "\n",
        "text_inputs = dict(\n",
        "    inputs_embeds=inputs_embeds,\n",
        "    position_ids=position_ids,\n",
        "    **dummy_past_key_values_kwargs,\n",
        ")\n",
        "text_inputs_positional = tuple(text_inputs.values())\n",
        "text_outputs = model.forward(*text_inputs_positional)  # Test forward pass\n",
        "\n",
        "## Vision inputs\n",
        "size = processor.image_processor.size\n",
        "w, h = size['width'], size['height']\n",
        "pixel_values = torch.randn(2, 3, h, w, requires_grad=True)\n",
        "vision_inputs = dict(pixel_values=pixel_values)\n",
        "vision_inputs_positional = tuple(vision_inputs.values())\n",
        "vision_outputs = vision_model.forward(*vision_inputs_positional)  # Test forward pass\n",
        "\n",
        "\n",
        "\n",
        "# ONNX Exports\n",
        "from torch.onnx._globals import GLOBALS\n",
        "GLOBALS.onnx_shape_inference = False # Bug in pytorch\n",
        "\n",
        "## Text model (Gemma 2).\n",
        "TEXT_MODEL_OUTPUT_PATH=os.path.join(TEMP_MODEL_OUTPUT_FOLDER, TEXT_MODEL_NAME)\n",
        "torch.onnx.export(\n",
        "    model,\n",
        "    args=text_inputs_positional,\n",
        "    f=TEXT_MODEL_OUTPUT_PATH,\n",
        "    export_params=True,\n",
        "    opset_version=14,\n",
        "    do_constant_folding=True,\n",
        "    input_names=list(text_inputs.keys()),\n",
        "    output_names=[\"logits\"]\n",
        "    + [f\"present.{i}.{key}\" for i in range(num_layers) for key in [\"key\", \"value\"]],\n",
        "    dynamic_axes={\n",
        "        \"inputs_embeds\": {0: \"batch_size\", 1: \"sequence_length\"},\n",
        "        \"position_ids\": {0: \"batch_size\", 1: \"sequence_length\"},\n",
        "        **{\n",
        "            f\"past_key_values.{i}.{key}\": {0: \"batch_size\", 2: \"past_sequence_length\"}\n",
        "            for i in range(num_layers)\n",
        "            for key in [\"key\", \"value\"]\n",
        "        },\n",
        "        \"logits\": {0: \"batch_size\", 1: \"sequence_length\"},\n",
        "        **{\n",
        "            f\"present.{i}.{key}\": {0: \"batch_size\", 2: \"total_sequence_length\"}\n",
        "            for i in range(num_layers)\n",
        "            for key in [\"key\", \"value\"]\n",
        "        },\n",
        "    },\n",
        "    external_data_format=True,\n",
        ")\n",
        "\n",
        "## Vision model (SigLIP).\n",
        "VISION_MODEL_OUTPUT_PATH = os.path.join(TEMP_MODEL_OUTPUT_FOLDER, VISION_MODEL_NAME)\n",
        "torch.onnx.export(\n",
        "    vision_model,\n",
        "    args=vision_inputs_positional,\n",
        "    f=VISION_MODEL_OUTPUT_PATH,\n",
        "    export_params=True,\n",
        "    opset_version=14,\n",
        "    do_constant_folding=True,\n",
        "    input_names=['pixel_values'],\n",
        "    output_names=['image_features'],\n",
        "    dynamic_axes={\n",
        "        'pixel_values': {0: 'batch_size'},\n",
        "        'image_features': {0: 'batch_size'}\n",
        "    },\n",
        ")\n",
        "\n",
        "input_ids = torch.randint(0, embed_layer.num_embeddings, (batch_size, sequence_length))\n",
        "\n",
        "## Embedding model\n",
        "EMBED_MODEL_OUTPUT_PATH = os.path.join(TEMP_MODEL_OUTPUT_FOLDER, EMBED_MODEL_NAME)\n",
        "torch.onnx.export(\n",
        "    embed_layer,\n",
        "    args=(input_ids,),\n",
        "    f=EMBED_MODEL_OUTPUT_PATH,\n",
        "    export_params=True,\n",
        "    opset_version=14,\n",
        "    do_constant_folding=True,\n",
        "    input_names=['input_ids'],\n",
        "    output_names=['inputs_embeds'],\n",
        "    dynamic_axes={\n",
        "        'input_ids': {0: 'batch_size', 1: 'sequence_length'},\n",
        "        'inputs_embeds': {0: 'batch_size', 1: 'sequence_length'}\n",
        "    },\n",
        ")\n",
        "\n",
        "\n",
        "# Post-processing\n",
        "import onnx\n",
        "import onnxslim\n",
        "from optimum.onnx.graph_transformations import check_and_save_model\n",
        "\n",
        "os.makedirs(FINAL_MODEL_OUTPUT_FOLDER, exist_ok=True)\n",
        "for name in (TEXT_MODEL_NAME, VISION_MODEL_NAME, EMBED_MODEL_NAME):\n",
        "    temp_model_path = os.path.join(TEMP_MODEL_OUTPUT_FOLDER, name)\n",
        "\n",
        "    onnx.shape_inference.infer_shapes_path(temp_model_path, check_type=True, strict_mode=True)\n",
        "\n",
        "    ## Attempt to optimize the model with onnxslim\n",
        "    \"\"\"\n",
        "    try:\n",
        "        onnx_model = onnxslim.slim(temp_model_path)\n",
        "    except Exception as e:\n",
        "        print(f\"Failed to slim {temp_model_path}: {e}\")\n",
        "        onnx_model = onnx.load(temp_model_path)\n",
        "    \"\"\"\n",
        "    onnx_model = onnx.load(temp_model_path)\n",
        "\n",
        "    ## Save model\n",
        "    final_model_path = os.path.join(FINAL_MODEL_OUTPUT_FOLDER, name)\n",
        "    check_and_save_model(onnx_model, final_model_path)\n",
        "\n",
        "\n",
        "# Minify tokenizer.json\n",
        "import json\n",
        "tokenizer_path = os.path.join(OUTPUT_FOLDER, \"tokenizer.json\")\n",
        "with open(tokenizer_path, \"r\") as f:\n",
        "    tokenizer = json.load(f)\n",
        "with open(tokenizer_path, \"w\") as f:\n",
        "    json.dump(tokenizer, f) # No need for indenting\n",
        "\n",
        "# Add head_dim and num_image_tokens to config.json\n",
        "config_path = os.path.join(OUTPUT_FOLDER, \"config.json\")\n",
        "with open(config_path, \"r\") as f:\n",
        "    config = json.load(f)\n",
        "config[\"text_config\"][\"head_dim\"] = head_dim\n",
        "config[\"num_image_tokens\"] = config[\"text_config\"][\"num_image_tokens\"]\n",
        "with open(config_path, \"w\") as f:\n",
        "    json.dump(config, f, indent=2)\n",
        "\n",
        "\n",
        "## Cleanup\n",
        "import shutil\n",
        "shutil.rmtree(TEMP_MODEL_OUTPUT_FOLDER)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FRIKmWypS6dC"
      },
      "source": [
        "### Step 4: Quantize the ONNX model weights (optional, but recommended)\n",
        "To optimize inference performance, it is recommended to quantize the ONNX model weights. We will be quantizing to the following precision data types:\n",
        "\n",
        "* fp16\n",
        "* int8\n",
        "* uint8\n",
        "* q4\n",
        "* q4f16\n",
        "* bnb4\n",
        "\n",
        "The overall quantization process will take approximately 40-45 minutes."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "m8gm8iklTV0X"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "--2025-02-19 19:11:50--  https://raw.githubusercontent.com/NSTiwari/PaliGemma2-ONNX-Transformers.js/main/quantize.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 12362 (12K) [text/plain]\n",
            "Saving to: ‘quantize.py’\n",
            "\n",
            "quantize.py         100%[===================>]  12.07K  --.-KB/s    in 0s      \n",
            "\n",
            "2025-02-19 19:11:51 (133 MB/s) - ‘quantize.py’ saved [12362/12362]\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Python script to quantize the ONNX model weights.\n",
        "!wget https://raw.githubusercontent.com/NSTiwari/PaliGemma2-ONNX-Transformers.js/main/quantize.py\n",
        "\n",
        "# Create a new directory to store quantized weights.\n",
        "!mkdir onnx_model_quantized"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FVo8RfNcUc2M"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[1;30;43mStreaming output truncated to the last 5000 lines.\u001b[0m\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Mul ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Softmax ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2011 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2013 ...\n",
            "2025-02-19 19:54:31,514 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:31,515 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:31,515 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:31,515 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:31,515 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/Add ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Add ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Div ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:31,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:31,533 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:31,576 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:31,576 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:31,576 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:31,576 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/Add_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Add ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Div ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Gather ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:31,594 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,598 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,598 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:31,598 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:31,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2073 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2075 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2080 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2082 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2087 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:31,609 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2089 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Slice ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Cast ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Div ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Mul ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Softmax ...\n",
            "2025-02-19 19:54:31,610 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2117 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2119 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:31,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/Add ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Add ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Div ...\n",
            "2025-02-19 19:54:31,615 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:31,616 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:31,616 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:31,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:31,629 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:31,629 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:31,629 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:31,629 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:31,629 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:31,677 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:31,677 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:31,677 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:31,677 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/Add_1 ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Add ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Div ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Gather ...\n",
            "2025-02-19 19:54:31,701 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:31,702 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:31,702 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:31,702 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:31,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,710 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,710 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:31,711 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2179 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2181 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2186 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2188 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2193 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2195 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:31,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Slice ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Cast ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Div ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Mul ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Softmax ...\n",
            "2025-02-19 19:54:31,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2223 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2225 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:31,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/Add ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Add ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Div ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:31,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:31,722 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:31,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:31,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:31,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:31,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:31,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:31,778 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:31,778 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:31,778 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:31,778 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/Add_1 ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Add ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Div ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant ...\n",
            "2025-02-19 19:54:31,803 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Gather ...\n",
            "2025-02-19 19:54:31,804 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:31,804 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:31,804 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:31,804 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:31,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,812 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,812 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:31,812 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2285 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2287 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2292 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2294 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2299 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2301 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:31,817 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Slice ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Cast ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Div ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Mul ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Softmax ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:31,818 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2329 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2331 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:31,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/Add ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Add ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Div ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:31,823 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:31,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:31,878 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:31,878 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:31,878 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:31,878 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,900 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,900 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/Add_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Add ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Div ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Gather ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:31,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,905 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:31,905 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:31,905 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:31,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:31,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2391 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2393 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2398 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2400 ...\n",
            "2025-02-19 19:54:31,914 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2405 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2407 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Slice ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Cast ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Div ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Mul ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:31,915 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Softmax ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2435 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2437 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:31,916 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,920 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/Add ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Add ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Div ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:31,921 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:31,934 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:31,979 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:31,979 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:31,979 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:31,979 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,997 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:31,997 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:31,997 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/Add_1 ...\n",
            "2025-02-19 19:54:31,997 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:31,997 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Add ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Div ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Gather ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:31,998 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,007 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,007 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,007 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2497 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2499 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2504 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2506 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2511 ...\n",
            "2025-02-19 19:54:32,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2513 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Div ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2541 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2543 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,013 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/Add ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,018 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,018 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,018 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,018 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,018 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,080 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,080 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,080 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,080 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/Add_1 ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,103 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,104 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2603 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2605 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2610 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2612 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2617 ...\n",
            "2025-02-19 19:54:32,117 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2619 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Div ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,118 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2647 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2649 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/Add ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,180 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,180 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,180 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,180 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/Add_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,205 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,211 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,211 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,211 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,218 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,218 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,218 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2709 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2711 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2716 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2718 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2723 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2725 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,224 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Div ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,225 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2753 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2755 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,226 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/Add ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,232 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,252 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,281 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,281 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,281 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,281 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/Add_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,305 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,312 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,312 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,312 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,318 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,318 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,318 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,324 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2815 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2817 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2822 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2824 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2829 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2831 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Div ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2859 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2861 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,327 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/Add ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,333 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,353 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,381 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,382 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,382 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,382 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,405 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,405 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,405 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/Add_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,406 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,412 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,412 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,412 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,418 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,418 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,418 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2921 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2923 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2928 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2930 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2935 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2937 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Div ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,426 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2965 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2967 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,427 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/Add ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,453 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,454 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,454 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,482 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,482 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,482 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,482 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/Add_1 ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,506 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,507 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,513 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,513 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,513 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,519 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,520 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3027 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3029 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3034 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3036 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3041 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3043 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Div ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3071 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3073 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,528 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/Add ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,534 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,535 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,535 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,554 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,554 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,554 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,554 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,554 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,583 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,583 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,583 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,583 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/Add_1 ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,606 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,607 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,613 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,613 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,613 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,619 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,620 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,620 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3133 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3135 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3140 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3142 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3147 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3149 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Div ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3177 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3179 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,628 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/Add ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,635 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,654 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,655 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,655 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,655 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,683 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,683 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,684 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,684 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,708 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,708 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,708 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/Add_1 ...\n",
            "2025-02-19 19:54:32,708 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,709 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,715 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,716 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,722 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,722 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,722 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3239 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3241 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3246 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3248 ...\n",
            "2025-02-19 19:54:32,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3253 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3255 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Div ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3283 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3285 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,730 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,736 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/Add ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,737 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,784 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,784 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,784 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,784 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/Add_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,808 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,814 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,814 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,814 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,821 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,821 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3345 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3347 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3352 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3354 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3359 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3361 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Div ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,828 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3389 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3391 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/Add ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,855 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,885 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,885 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,885 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,885 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/Add_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Add ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Div ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Gather ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:32,910 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,917 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:32,917 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:32,917 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,923 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:32,923 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:32,923 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3451 ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3453 ...\n",
            "2025-02-19 19:54:32,929 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3458 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3460 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3465 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3467 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Slice ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Cast ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Div ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Mul ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:32,930 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Softmax ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3495 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3497 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:32,931 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:32,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/Add ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Add ...\n",
            "2025-02-19 19:54:32,937 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:32,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Div ...\n",
            "2025-02-19 19:54:32,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:32,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:32,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,957 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:32,957 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:32,957 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:32,958 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:32,985 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:32,985 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:32,985 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:32,986 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/Add_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Add ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Div ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Gather ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:33,011 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:33,012 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:33,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:33,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:33,017 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:33,023 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:33,023 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:33,023 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:33,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:33,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:33,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3557 ...\n",
            "2025-02-19 19:54:33,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:33,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3559 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3564 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3566 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3571 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3573 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Slice ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Cast ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Div ...\n",
            "2025-02-19 19:54:33,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Mul ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Softmax ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:33,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3601 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3603 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:33,031 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/Add ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Add ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Div ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:33,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:33,057 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:33,086 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:33,086 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:33,086 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:33,086 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:33,112 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:33,112 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:33,112 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/Add_1 ...\n",
            "2025-02-19 19:54:33,112 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/ReduceMean ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Sub ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Constant ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Pow ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/ReduceMean_1 ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Constant_1 ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Add ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Sqrt ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Div ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Mul ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Add_1 ...\n",
            "2025-02-19 19:54:33,113 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /multi_modal_projector/linear/MatMul ...\n",
            "2025-02-19 19:54:33,122 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /multi_modal_projector/linear/MatMul ...\n",
            "2025-02-19 19:54:33,122 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /multi_modal_projector/linear/Add ...\n",
            "2025-02-19 19:54:33,122 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Constant ...\n",
            "2025-02-19 19:54:33,122 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Div ...\n",
            "\n",
            " - Quantizing to q4:  67% 4/6 [03:51<01:36, 48.07s/it]\u001b[A\n",
            " - Quantizing to q4f16:  67% 4/6 [03:51<01:36, 48.07s/it]\u001b[A2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Cast ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/patch_embedding/Conv ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Shape ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Constant ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Constant_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Constant_2 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Slice ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Constant_3 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Concat ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Reshape ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Transpose ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/position_embedding/Constant ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/position_embedding/Gather ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/embeddings/Add ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Add ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Div ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Shape ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Gather ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:35,555 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:35,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,560 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,560 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:35,560 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,564 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,564 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:35,564 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,568 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,568 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:35,568 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_801 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_803 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Concat ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_808 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_810 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_815 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_817 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Reshape ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Transpose ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Slice ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Cast ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Div ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:35,569 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Mul ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Softmax ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_845 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_847 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:35,570 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,574 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,574 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:35,574 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/Add ...\n",
            "2025-02-19 19:54:35,574 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:35,574 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Add ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Div ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:35,575 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,589 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,589 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:35,589 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:35,592 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:35,593 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.0/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.0/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:35,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.0/Add_1 ...\n",
            "2025-02-19 19:54:35,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:35,611 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Add ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Div ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Shape ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Gather ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:35,612 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,616 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,616 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:35,616 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,621 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,621 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:35,621 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_907 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_909 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Concat ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_914 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_916 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_921 ...\n",
            "2025-02-19 19:54:35,625 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_923 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Reshape ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Transpose ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Slice ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Cast ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Div ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Mul ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Softmax ...\n",
            "2025-02-19 19:54:35,626 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_951 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_953 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:35,627 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/Add ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Add ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Div ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:35,632 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:35,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:35,693 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:35,693 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:35,693 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.1/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.1/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.1/Add_1 ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:35,712 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Add ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Div ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Shape ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Gather ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:35,713 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:35,717 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,721 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:35,722 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1013 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1015 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Concat ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1020 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1022 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1027 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1029 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Reshape ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Transpose ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:35,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Slice ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Cast ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Div ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Mul ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Softmax ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1057 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1059 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:35,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/Add ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Add ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Div ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:35,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:35,747 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:35,748 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:35,794 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:35,794 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:35,794 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.2/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.2/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.2/Add_1 ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Add ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Div ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Shape ...\n",
            "2025-02-19 19:54:35,819 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant ...\n",
            "2025-02-19 19:54:35,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Gather ...\n",
            "2025-02-19 19:54:35,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:35,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:35,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:35,820 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,824 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,824 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:35,824 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:35,829 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,833 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,833 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1119 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1121 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Concat ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1126 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1128 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1133 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1135 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Reshape ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Transpose ...\n",
            "2025-02-19 19:54:35,834 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Slice ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Cast ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Div ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Mul ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Softmax ...\n",
            "2025-02-19 19:54:35,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1163 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1165 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:35,836 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/Add ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Add ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Div ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:35,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:35,854 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:35,895 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:35,895 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:35,895 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.3/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.3/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.3/Add_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Add ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Div ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Shape ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Gather ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:35,913 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,918 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:35,918 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:35,918 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,922 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:35,922 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:35,922 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1225 ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1227 ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:35,926 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Concat ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1232 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1234 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1239 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1241 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Reshape ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Transpose ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Slice ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Cast ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Div ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Mul ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:35,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/MatMul ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Softmax ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1269 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1271 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:35,928 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/Add ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:35,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Add ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Div ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:35,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,945 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:35,945 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:35,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:35,995 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:35,995 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:35,995 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.4/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.4/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.4/Add_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,021 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,026 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,026 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,026 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,030 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1331 ...\n",
            "2025-02-19 19:54:36,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1333 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1338 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1340 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1345 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1347 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,035 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Div ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,036 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1375 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1377 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,041 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,041 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,041 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/Add ...\n",
            "2025-02-19 19:54:36,041 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,055 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,096 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,096 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,096 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.5/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.5/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.5/Add_1 ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,114 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,115 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,119 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,124 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,124 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,124 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1437 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1439 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1444 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1446 ...\n",
            "2025-02-19 19:54:36,128 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1451 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1453 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Div ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,129 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1481 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1483 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/Add ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,148 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,148 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,148 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,149 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,197 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,197 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,197 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.6/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.6/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.6/Add_1 ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,222 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,223 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,227 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,227 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,227 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,231 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,231 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,231 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1543 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1545 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1550 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1552 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1557 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1559 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,236 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Div ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1587 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1589 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/Add ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,255 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,297 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,297 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,297 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.7/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.7/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.7/Add_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,320 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,320 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,321 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,325 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1649 ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1651 ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,329 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1656 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1658 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1663 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1665 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Div ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,330 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1693 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1695 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/Add ...\n",
            "2025-02-19 19:54:36,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,348 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,349 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,398 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,398 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,398 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.8/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.8/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.8/Add_1 ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,424 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,425 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,429 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,429 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,429 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,433 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,434 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1755 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1757 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1762 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1764 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,438 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1769 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1771 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Div ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1799 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1801 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,440 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,441 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,441 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/Add ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,447 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,460 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,461 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,499 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,499 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,499 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.9/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,516 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.9/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,516 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.9/Add_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,517 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,521 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,521 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,521 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,526 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1861 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1863 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1868 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1870 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1875 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1877 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,530 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Div ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,531 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1905 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1907 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/Add ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,550 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,599 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,599 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,599 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,599 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.10/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.10/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.10/Add_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,623 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,630 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,630 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,630 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,634 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1967 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1969 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1974 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1976 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1981 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_1983 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,639 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Div ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2011 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2013 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/Add ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,658 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,700 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,700 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,700 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,700 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.11/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.11/Add_1 ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,718 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,719 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,723 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,723 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,723 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,728 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2073 ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2075 ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,732 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2080 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2082 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2087 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2089 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Div ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,734 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2117 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2119 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,735 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/Add ...\n",
            "2025-02-19 19:54:36,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,753 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,800 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,801 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,801 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,801 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.12/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.12/Add_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,826 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,827 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,831 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,831 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,831 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,835 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2179 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2181 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2186 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2188 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2193 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2195 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,840 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Div ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2223 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2225 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,842 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/Add ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,859 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,860 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,860 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,860 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,860 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:36,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:36,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:36,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:36,901 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.13/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.13/Add_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Add ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Div ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Gather ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:36,927 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:36,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:36,932 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,936 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:36,936 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:36,936 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,940 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2285 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2287 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2292 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2294 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2299 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2301 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:36,941 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Slice ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Cast ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Div ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Mul ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:36,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Softmax ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2329 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2331 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:36,943 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/Add ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Add ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Div ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:36,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:36,961 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:36,962 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:36,962 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:36,962 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,002 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.14/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.14/Add_1 ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,028 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,029 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,037 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,038 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,038 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2391 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2393 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2398 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2400 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2405 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2407 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Div ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,043 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2435 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,044 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2437 ...\n",
            "2025-02-19 19:54:37,045 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,045 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,045 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,045 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,045 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/Add ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,050 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,050 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,062 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,102 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,102 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,102 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,102 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,120 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.15/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,120 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,120 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.15/Add_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,121 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,125 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,125 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,125 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,130 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2497 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2499 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2504 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2506 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,134 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2511 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2513 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Div ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,135 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2541 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2543 ...\n",
            "2025-02-19 19:54:37,137 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,137 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,137 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,137 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,137 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/Add ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,153 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,153 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,153 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,154 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,203 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,203 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,203 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,203 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.16/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.16/Add_1 ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,228 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,229 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,233 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,233 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,233 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,237 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2603 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2605 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2610 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2612 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2617 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2619 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,242 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Div ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2647 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2649 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,244 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/Add ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,248 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,261 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,303 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,304 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,304 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,304 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,321 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.17/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.17/Add_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,322 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,326 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,331 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2709 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2711 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2716 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2718 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2723 ...\n",
            "2025-02-19 19:54:37,335 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2725 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Div ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,336 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2753 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2755 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,337 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/Add ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,341 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,342 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,354 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,355 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,404 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,404 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,404 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,404 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.18/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.18/Add_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,430 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,435 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,435 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,435 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,439 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,443 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,443 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,443 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2815 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2817 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2822 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2824 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2829 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2831 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Div ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,444 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2859 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2861 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,445 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/Add ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,450 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,463 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,505 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,505 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,505 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,505 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.19/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.19/Add_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,523 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,527 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,532 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2921 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2923 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2928 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2930 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2935 ...\n",
            "2025-02-19 19:54:37,536 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2937 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Div ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,537 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2965 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_2967 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,538 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/Add ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,542 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,543 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,556 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,605 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.20/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.20/Add_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,631 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,636 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,636 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,636 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,640 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3027 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3029 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3034 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3036 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3041 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3043 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,644 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Div ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,645 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3071 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3073 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,646 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/Add ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,650 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,663 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,706 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.21/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.21/Add_1 ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,724 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,725 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,729 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,733 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3133 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3135 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3140 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3142 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3147 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3149 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,738 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Div ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,739 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3177 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3179 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,740 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/Add ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,744 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,757 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,806 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,806 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,806 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,806 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.22/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.22/Add_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,832 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,837 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,837 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,837 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,841 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,845 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,845 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,845 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3239 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3241 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3246 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3248 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3253 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3255 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Div ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,846 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3283 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3285 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,847 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/Add ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,852 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,865 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:37,907 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:37,907 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:37,907 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:37,907 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.23/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.23/Add_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Add ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Div ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Gather ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:37,933 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:37,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:37,938 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:37,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:37,942 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3345 ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3347 ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat ...\n",
            "2025-02-19 19:54:37,946 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3352 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3354 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3359 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3361 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Slice ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Cast ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Div ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Mul ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:37,947 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Softmax ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3389 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3391 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:37,948 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,952 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:37,952 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:37,952 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/Add ...\n",
            "2025-02-19 19:54:37,952 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Add ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Div ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:37,953 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:37,966 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:38,008 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:38,008 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:38,008 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:38,008 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.24/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.24/Add_1 ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Add ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Div ...\n",
            "2025-02-19 19:54:38,033 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Gather ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:38,034 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:38,038 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:38,038 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:38,038 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:38,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:38,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:38,042 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3451 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3453 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3458 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3460 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3465 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3467 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:38,047 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Slice ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Cast ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Div ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Mul ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Softmax ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:38,048 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3495 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3497 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:38,049 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/Add ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Add ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Div ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:38,053 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:38,066 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:38,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:38,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:38,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:38,108 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.25/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:38,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.25/Add_1 ...\n",
            "2025-02-19 19:54:38,126 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/ReduceMean ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Sub ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Constant ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Pow ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/ReduceMean_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Constant_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Add ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Sqrt ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Div ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Mul ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm1/Add_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Gather ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Gather_1 ...\n",
            "2025-02-19 19:54:38,127 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:38,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/MatMul ...\n",
            "2025-02-19 19:54:38,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/q_proj/Add ...\n",
            "2025-02-19 19:54:38,131 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:38,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/MatMul ...\n",
            "2025-02-19 19:54:38,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/k_proj/Add ...\n",
            "2025-02-19 19:54:38,136 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/MatMul ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/v_proj/Add ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3557 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3559 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_1 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_2 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_3 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3564 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_2 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3566 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_3 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_4 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_5 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_1 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3571 ...\n",
            "2025-02-19 19:54:38,140 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_4 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3573 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_5 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_6 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_7 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_2 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_1 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_2 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_1 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Shape_2 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_8 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_9 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Slice ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Cast ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_10 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Div ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Cast_1 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_2 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt_1 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Mul ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Sqrt_2 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Mul_1 ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Softmax ...\n",
            "2025-02-19 19:54:38,141 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - MatMul doesn't have const weight. Skip to quantize\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/MatMul_1 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Transpose_3 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3601 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_6 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize Constant_3603 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Unsqueeze_7 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Constant_11 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Concat_3 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/Reshape_3 ...\n",
            "2025-02-19 19:54:38,142 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/MatMul ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/self_attn/out_proj/Add ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/Add ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/ReduceMean ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Sub ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Constant ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Pow ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/ReduceMean_1 ...\n",
            "2025-02-19 19:54:38,146 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Constant_1 ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Add ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Sqrt ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Div ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Mul ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/layer_norm2/Add_1 ...\n",
            "2025-02-19 19:54:38,147 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/mlp/fc1/MatMul ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc1/Add ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_1 ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant ...\n",
            "2025-02-19 19:54:38,159 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_2 ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Add ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_1 ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_3 ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Tanh ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_2 ...\n",
            "2025-02-19 19:54:38,160 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Add_1 ...\n",
            "2025-02-19 19:54:38,209 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_4 ...\n",
            "2025-02-19 19:54:38,209 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Constant_3 ...\n",
            "2025-02-19 19:54:38,209 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/activation_fn/Mul_5 ...\n",
            "2025-02-19 19:54:38,209 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /vision_tower/vision_model/encoder/layers.26/mlp/fc2/MatMul ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/mlp/fc2/Add ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/encoder/layers.26/Add_1 ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/ReduceMean ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Sub ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Constant ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Pow ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/ReduceMean_1 ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Constant_1 ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Add ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Sqrt ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Div ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Mul ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /vision_tower/vision_model/post_layernorm/Add_1 ...\n",
            "2025-02-19 19:54:38,235 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - start to quantize /multi_modal_projector/linear/MatMul ...\n",
            "2025-02-19 19:54:38,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - complete quantization of /multi_modal_projector/linear/MatMul ...\n",
            "2025-02-19 19:54:38,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /multi_modal_projector/linear/Add ...\n",
            "2025-02-19 19:54:38,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Constant ...\n",
            "2025-02-19 19:54:38,243 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Div ...\n",
            "/usr/local/lib/python3.11/dist-packages/onnxconverter_common/float16.py:43: UserWarning: the float32 number 4.085620730620576e-14 will be truncated to 1e-07\n",
            "  warnings.warn(\"the float32 number {} will be truncated to {}\".format(pos_min, min_positive_val))\n",
            "/usr/local/lib/python3.11/dist-packages/onnxconverter_common/float16.py:53: UserWarning: the float32 number -4.6407322429331543e-14 will be truncated to -1e-07\n",
            "  warnings.warn(\"the float32 number {} will be truncated to {}\".format(neg_max, -min_positive_val))\n",
            "\n",
            " - Quantizing to q4f16:  83% 5/6 [04:03<00:35, 35.01s/it]\u001b[A\n",
            " - Quantizing to bnb4:  83% 5/6 [04:03<00:35, 35.01s/it] \u001b[A\n",
            " - Quantizing to bnb4: 100% 6/6 [04:08<00:00, 41.44s/it]\n",
            "Processing output/google/paligemma2-3b-mix-224/onnx/embed_tokens.onnx:  67% 2/3 [42:49<18:21, 1102.00s/it]  \n",
            "  0% 0/6 [00:00<?, ?it/s]\u001b[A\n",
            " - Quantizing to fp16:   0% 0/6 [00:00<?, ?it/s]\u001b[A\n",
            " - Quantizing to fp16:  17% 1/6 [02:24<12:00, 144.07s/it]\u001b[A\n",
            " - Quantizing to int8:  17% 1/6 [02:24<12:00, 144.07s/it]\u001b[A\n",
            " - Quantizing to int8:  33% 2/6 [04:14<08:18, 124.56s/it]\u001b[A\n",
            " - Quantizing to uint8:  33% 2/6 [04:14<08:18, 124.56s/it]\u001b[A\n",
            " - Quantizing to uint8:  50% 3/6 [06:17<06:11, 123.77s/it]\u001b[A\n",
            " - Quantizing to q4:  50% 3/6 [06:17<06:11, 123.77s/it]   \u001b[A2025-02-19 20:01:10,316 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Gather ...\n",
            "\n",
            " - Quantizing to q4:  67% 4/6 [07:50<03:42, 111.32s/it]\u001b[A\n",
            " - Quantizing to q4f16:  67% 4/6 [07:50<03:42, 111.32s/it]\u001b[A2025-02-19 20:02:42,552 onnxruntime.quantization.matmul_4bits_quantizer [INFO] - skip to quantize /Gather ...\n",
            "\n",
            " - Quantizing to q4f16:  83% 5/6 [09:52<01:55, 115.26s/it]\u001b[A\n",
            " - Quantizing to bnb4:  83% 5/6 [09:52<01:55, 115.26s/it] \u001b[A\n",
            " - Quantizing to bnb4: 100% 6/6 [11:32<00:00, 115.35s/it]\n",
            "Processing output/google/paligemma2-3b-mix-224/onnx/embed_tokens.onnx: 100% 3/3 [54:21<00:00, 1087.28s/it]\n"
          ]
        }
      ],
      "source": [
        "!python quantize.py \\\n",
        "  --input_folder $FINAL_MODEL_OUTPUT_FOLDER \\\n",
        "  --output_folder onnx_model_quantized \\\n",
        "  --modes fp16 int8 uint8 q4 q4f16 bnb4 \\\n",
        "  --per_channel \\\n",
        "  --reduce_range \\\n",
        "  --block_size 64 \\\n",
        "  --is_symmetric \\\n",
        "  --accuracy_level 2 \\\n",
        "  --quant_type 1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "yag_L-yYZWfj"
      },
      "outputs": [],
      "source": [
        "# Copy the quantized ONNX weights to the final model output folder.\n",
        "source = \"/content/onnx_model_quantized/.\"\n",
        "destination = f\"/content/output/{model_id}/onnx/\"\n",
        "\n",
        "!cp -a $source $destination"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pP2Fu2n2aL_R"
      },
      "source": [
        "### Step 5: Upload the ONNX weights on Hugging Face"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QdflTumaaQ3W"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "18b379ce846240b681c945f3bf912204",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_fp16.onnx:   0%|          | 0.00/1.83M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "24539377e7564ef89f18ef788423781e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged.onnx:   0%|          | 0.00/1.81M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0fec50109cb144fd99b2ee8f730dc6d4",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Upload 29 LFS files:   0%|          | 0/29 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "2a8bc6947801434aa88b0bae83134ed6",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_bnb4.onnx:   0%|          | 0.00/1.47G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0c3a7f5e5b604b0887f44f3959b06bd5",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged.onnx_data:   0%|          | 0.00/10.5G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "104981fb1611460296807fd9f9507836",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_fp16.onnx_data:   0%|          | 0.00/5.23G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7f5a54533ada46baabb2e9f1607bb9b6",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_int8.onnx:   0%|          | 0.00/6.83M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4b37934b1fed4435aace4c9d06457942",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_int8.onnx_data:   0%|          | 0.00/2.62G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "237369418ae6449b93dabc7e7d37b8c7",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_q4.onnx:   0%|          | 0.00/1.47G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "3f0df2b7fc6545a58b82a8e3cdd30747",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_q4f16.onnx:   0%|          | 0.00/1.39G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8bf4b2946dff4c608d0912309b6ab4a1",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_uint8.onnx:   0%|          | 0.00/6.83M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "62bdba3828664c87be3f319993d1997c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "decoder_model_merged_uint8.onnx_data:   0%|          | 0.00/2.62G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ad329d7412d04e3eb110629d86ccf1b2",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens.onnx:   0%|          | 0.00/299 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ab32358d58004008ab23cd52e44ec8ba",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_q4.onnx_data:   0%|          | 0.00/2.37G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4d611e27fd5841cb81e0ac23a370ca72",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_bnb4.onnx:   0%|          | 0.00/323 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0c603ed29a9a4962908dfb03c1a5286c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_q4.onnx_data:   0%|          | 0.00/2.37G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d67fd537336341d4bed99f7d94e8a35a",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_fp16.onnx:   0%|          | 0.00/1.19G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b64e7bf415274561b6599db7273483c1",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_uint8.onnx:   0%|          | 0.00/593M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "1dfe333ffc904142ab7b59efd4c10b9e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_q4.onnx:   0%|          | 0.00/321 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "89bc68e58dd6443c8df15356154a0797",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_q4.onnx_data:   0%|          | 0.00/2.37G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "faec4c54935b4fbba84fdf4443d19105",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_q4f16.onnx:   0%|          | 0.00/1.19G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ee6490f7713c4e8ea526d7428cd5ce60",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "embed_tokens_uint8.onnx:   0%|          | 0.00/593M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "550c39c87c22483ca03b42c9b657a80e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder.onnx:   0%|          | 0.00/1.66G [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b1829c37a5c84d15b30745761268eb39",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_bnb4.onnx:   0%|          | 0.00/239M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "a0eacf4cca8e4756adaf0ccf421369dd",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_fp16.onnx:   0%|          | 0.00/831M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "c5d171d4686a438a98b9f50f4c5eee79",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_int8.onnx:   0%|          | 0.00/419M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "86ac5cf440624ca2bbc61ce50afb810e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_q4.onnx:   0%|          | 0.00/240M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "99519203f95f49b2b9be20e188a575aa",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_q4f16.onnx:   0%|          | 0.00/224M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "1be9d7befdcf4ffc9a3f9cdd0126b89f",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vision_encoder_uint8.onnx:   0%|          | 0.00/419M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "6247f1bc1faa4e708adfb7a51aaafd7f",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer.json:   0%|          | 0.00/18.7M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "CommitInfo(commit_url='https://huggingface.co/NSTiwari/paligemma2-3b-mix-224-onnx/commit/fb4873c575fbf05f2cbc813bcd524bf06f81b0e7', commit_message='google/paligemma2-3b-mix-224 ONNX', commit_description='', oid='fb4873c575fbf05f2cbc813bcd524bf06f81b0e7', pr_url=None, repo_url=RepoUrl('https://huggingface.co/NSTiwari/paligemma2-3b-mix-224-onnx', endpoint='https://huggingface.co', repo_type='model', repo_id='NSTiwari/paligemma2-3b-mix-224-onnx'), pr_revision=None, pr_num=None)"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from huggingface_hub import whoami\n",
        "from pathlib import Path\n",
        "from huggingface_hub import upload_folder, create_repo\n",
        "\n",
        "# Output directory.\n",
        "output_dir = f\"/content/output/{model_id}/\"\n",
        "username = whoami(token=Path(\"/root/.cache/huggingface/\"))[\"name\"]\n",
        "repo_id = f\"{username}/paligemma2-3b-mix-224-onnx\"\n",
        "\n",
        "repo_id = create_repo(repo_id, exist_ok=True).repo_id\n",
        "\n",
        "upload_folder(\n",
        "    repo_id=repo_id,\n",
        "    folder_path=output_dir,\n",
        "    commit_message=f\"{model_id} ONNX\",\n",
        "    ignore_patterns=[\"step_*\", \"epoch_*\"],\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nl_Pw0lVb6dy"
      },
      "source": [
        "Congratulations, we have successfully converted and quantized the PaliGemma 2 model to the ONNX format, making it compatible with 🤗 Transformers.js for inference on the web.\n",
        "\n",
        "Next, to run inference with the converted PaliGemma 2 ONNX model, refer to this [notebook](https://github.com/google-gemini/gemma-cookbook/blob/main/PaliGemma/[PaliGemma_2]Inference_PaliGemma2_with_Transformers_js.ipynb). For the web application, check out this [demo app](https://github.com/google-gemini/gemma-cookbook/tree/main/Demos/PaliGemma2-on-Web)."
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "[PaliGemma_2]Convert_PaliGemma2_to_ONNX.ipynb",
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}