specification/ai/data-plane/ModelInference/openapi/2025-04-01/openapi.yaml

openapi: 3.0.0
info:
  title: AI Model Inference
  version: '2025-04-01'
tags: []
paths:
  /chat/completions:
    post:
      operationId: getChatCompletions
      description: |-
        Gets chat completions for the provided chat messages.
        Completions support a wide variety of tasks and generate text that continues from or "completes"
        provided prompt data. The method makes a REST API call to the `/chat/completions` route
        on the given endpoint.
      parameters:
        - $ref: '#/components/parameters/Azure.Core.Foundations.ApiVersionParameter'
        - name: extra-parameters
          in: header
          required: false
          description: |-
            Controls what happens if extra parameters, undefined by the REST API,
            are passed in the JSON request payload.
            This sets the HTTP request header `extra-parameters`.
          schema:
            $ref: '#/components/schemas/ExtraParameters'
      responses:
        '200':
          description: The request has succeeded.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletions'
        default:
          description: An unexpected error response.
          headers:
            x-ms-error-code:
              required: false
              description: String error code indicating what went wrong.
              schema:
                type: string
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Azure.Core.Foundations.ErrorResponse'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ChatCompletionsOptions'
        description: The parameters of the chat completions request.
  /embeddings:
    post:
      operationId: getEmbeddings
      description: |-
        Return the embedding vectors for given text prompts.
        The method makes a REST API call to the `/embeddings` route on the given endpoint.
      parameters:
        - $ref: '#/components/parameters/Azure.Core.Foundations.ApiVersionParameter'
        - name: extra-parameters
          in: header
          required: false
          description: |-
            Controls what happens if extra parameters, undefined by the REST API,
            are passed in the JSON request payload.
            This sets the HTTP request header `extra-parameters`.
          schema:
            $ref: '#/components/schemas/ExtraParameters'
      responses:
        '200':
          description: The request has succeeded.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/EmbeddingsResult'
        default:
          description: An unexpected error response.
          headers:
            x-ms-error-code:
              required: false
              description: String error code indicating what went wrong.
              schema:
                type: string
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Azure.Core.Foundations.ErrorResponse'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/EmbeddingsOptions'
        description: The parameters of the embeddings request.
  /images/embeddings:
    post:
      operationId: getImageEmbeddings
      description: |-
        Return the embedding vectors for given images.
        The method makes a REST API call to the `/images/embeddings` route on the given endpoint.
      parameters:
        - $ref: '#/components/parameters/Azure.Core.Foundations.ApiVersionParameter'
        - name: extra-parameters
          in: header
          required: false
          description: |-
            Controls what happens if extra parameters, undefined by the REST API,
            are passed in the JSON request payload.
            This sets the HTTP request header `extra-parameters`.
          schema:
            $ref: '#/components/schemas/ExtraParameters'
      responses:
        '200':
          description: The request has succeeded.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/EmbeddingsResult'
        default:
          description: An unexpected error response.
          headers:
            x-ms-error-code:
              required: false
              description: String error code indicating what went wrong.
              schema:
                type: string
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Azure.Core.Foundations.ErrorResponse'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ImageEmbeddingsOptions'
        description: The parameters of the image embeddings request.
  /info:
    get:
      operationId: getModelInfo
      description: |-
        Returns information about the AI model deployed.
        The method makes a REST API call to the `/info` route on the given endpoint.
        This method will only work when using Serverless API, Managed Compute, or Model .
        inference endpoint. Azure OpenAI endpoints don't support i.
      parameters:
        - $ref: '#/components/parameters/Azure.Core.Foundations.ApiVersionParameter'
        - name: model
          in: query
          required: false
          description: The model deployment name you want information from.
          schema:
            type: string
          explode: false
      responses:
        '200':
          description: The request has succeeded.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ModelInfo'
        default:
          description: An unexpected error response.
          headers:
            x-ms-error-code:
              required: false
              description: String error code indicating what went wrong.
              schema:
                type: string
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Azure.Core.Foundations.ErrorResponse'
security:
  - ApiKeyAuth: []
  - BearerAuth: []
  - OAuth2Auth:
      - https://cognitiveservices.azure.com/.default
components:
  parameters:
    Azure.Core.Foundations.ApiVersionParameter:
      name: api-version
      in: query
      required: true
      description: The API version to use for this operation.
      schema:
        type: string
        minLength: 1
      explode: false
  schemas:
    AudioContentFormat:
      anyOf:
        - type: string
        - type: string
          enum:
            - wav
            - mp3
      description: A representation of the possible audio formats for audio.
    Azure.Core.Foundations.Error:
      type: object
      required:
        - code
        - message
      properties:
        code:
          type: string
          description: One of a server-defined set of error codes.
        message:
          type: string
          description: A human-readable representation of the error.
        target:
          type: string
          description: The target of the error.
        details:
          type: array
          items:
            $ref: '#/components/schemas/Azure.Core.Foundations.Error'
          description: An array of details about specific errors that led to this reported error.
        innererror:
          allOf:
            - $ref: '#/components/schemas/Azure.Core.Foundations.InnerError'
          description: An object containing more specific information than the current object about the error.
      description: The error object.
    Azure.Core.Foundations.ErrorResponse:
      type: object
      required:
        - error
      properties:
        error:
          allOf:
            - $ref: '#/components/schemas/Azure.Core.Foundations.Error'
          description: The error object.
      description: A response containing error details.
    Azure.Core.Foundations.InnerError:
      type: object
      properties:
        code:
          type: string
          description: One of a server-defined set of error codes.
        innererror:
          allOf:
            - $ref: '#/components/schemas/Azure.Core.Foundations.InnerError'
          description: Inner error.
      description: An object containing more specific information about the error. As per Microsoft One API guidelines - https://github.com/microsoft/api-guidelines/blob/vNext/azure/Guidelines.md#handling-errors.
    ChatChoice:
      type: object
      required:
        - index
        - finish_reason
        - message
      properties:
        index:
          type: integer
          format: int32
          description: The ordered index associated with this chat completions choice.
        finish_reason:
          allOf:
            - $ref: '#/components/schemas/CompletionsFinishReason'
          nullable: true
          description: The reason that this chat completions choice completed its generated.
          readOnly: true
        message:
          allOf:
            - $ref: '#/components/schemas/ChatResponseMessage'
          description: The chat message for a given chat completions prompt.
          readOnly: true
      description: |-
        The representation of a single prompt completion as part of an overall chat completions request.
        Generally, `n` choices are generated per provided prompt with a default value of 1.
        Token limits and other settings may limit the number of choices generated.
    ChatCompletions:
      type: object
      required:
        - id
        - object
        - created
        - model
        - choices
        - usage
      properties:
        id:
          type: string
          description: A unique identifier associated with this chat completions response.
        object:
          type: string
          enum:
            - chat.completion
          description: The response object type, which is always `chat.completion`.
        created:
          type: integer
          format: unixtime
          description: |-
            The first timestamp associated with generation activity for this completions response,
            represented as seconds since the beginning of the Unix epoch of 00:00 on 1 Jan 1970.
          readOnly: true
        model:
          type: string
          description: The model used for the chat completion.
          readOnly: true
        choices:
          type: array
          items:
            $ref: '#/components/schemas/ChatChoice'
          minItems: 1
          description: |-
            The collection of completions choices associated with this completions response.
            Generally, `n` choices are generated per provided prompt with a default value of 1.
            Token limits and other settings may limit the number of choices generated.
          readOnly: true
        usage:
          allOf:
            - $ref: '#/components/schemas/CompletionsUsage'
          description: '  Usage information for tokens processed and generated as part of this completions operation.'
          readOnly: true
      description: |-
        Representation of the response data from a chat completions request.
        Completions support a wide variety of tasks and generate text that continues from or "completes"
        provided prompt data.
    ChatCompletionsAudio:
      type: object
      required:
        - id
        - expires_at
        - data
        - transcript
      properties:
        id:
          type: string
          description: |2-
              Unique identifier for the audio response. This value can be used in chat history messages instead of passing 
              the full audio object.
          readOnly: true
        expires_at:
          type: integer
          format: unixtime
          description: |-
            The Unix timestamp (in seconds) at which the audio piece expires and can't be any longer referenced by its ID in 
            multi-turn conversations.
          readOnly: true
        data:
          type: string
          description: Base64 encoded audio data
          readOnly: true
        format:
          allOf:
            - $ref: '#/components/schemas/AudioContentFormat'
          description: |-
            The format of the audio content. If format is not provided, it will match the format used in the
            input audio request.
          readOnly: true
        transcript:
          type: string
          description: The transcript of the audio file.
          readOnly: true
      description: A representation of the audio generated by the model.
    ChatCompletionsModality:
      anyOf:
        - type: string
        - type: string
          enum:
            - text
            - audio
      description: The modalities that the model is allowed to use for the chat completions response.
    ChatCompletionsNamedToolChoice:
      type: object
      required:
        - type
        - function
      properties:
        type:
          type: string
          enum:
            - function
          description: The type of the tool. Currently, only `function` is supported.
        function:
          allOf:
            - $ref: '#/components/schemas/ChatCompletionsNamedToolChoiceFunction'
          description: The function that should be called.
      description: A tool selection of a specific, named function tool that will limit chat completions to using the named function.
    ChatCompletionsNamedToolChoiceFunction:
      type: object
      required:
        - name
      properties:
        name:
          type: string
          description: The name of the function that should be called.
      description: A tool selection of a specific, named function tool that will limit chat completions to using the named function.
    ChatCompletionsOptions:
      type: object
      required:
        - messages
      properties:
        messages:
          type: array
          items:
            $ref: '#/components/schemas/ChatRequestMessage'
          minItems: 1
          description: |-
            The collection of context messages associated with this chat completions request.
            Typical usage begins with a chat message for the System role that provides instructions for
            the behavior of the assistant, followed by alternating messages between the User and
            Assistant roles.
        frequency_penalty:
          type: number
          format: float
          minimum: -2
          maximum: 2
          description: |-
            A value that influences the probability of generated tokens appearing based on their cumulative
            frequency in generated text.
            Positive values will make tokens less likely to appear as their frequency increases and
            decrease the likelihood of the model repeating the same statements verbatim.
            Supported range is [-2, 2].
          default: 0
        stream:
          type: boolean
          description: A value indicating whether chat completions should be streamed for this request.
        presence_penalty:
          type: number
          format: float
          minimum: -2
          maximum: 2
          description: |-
            A value that influences the probability of generated tokens appearing based on their existing
            presence in generated text.
            Positive values will make tokens less likely to appear when they already exist and increase the
            model's likelihood to output new topics.
            Supported range is [-2, 2].
          default: 0
        temperature:
          type: number
          format: float
          minimum: 0
          maximum: 1
          description: |-
            The sampling temperature to use that controls the apparent creativity of generated completions.
            Higher values will make output more random while lower values will make results more focused
            and deterministic.
            It is not recommended to modify temperature and top_p for the same completions request as the
            interaction of these two settings is difficult to predict.
            Supported range is [0, 1].
          default: 0.7
        top_p:
          type: number
          format: float
          minimum: 0
          maximum: 1
          description: |-
            An alternative to sampling with temperature called nucleus sampling. This value causes the
            model to consider the results of tokens with the provided probability mass. As an example, a
            value of 0.15 will cause only the tokens comprising the top 15% of probability mass to be
            considered.
            It is not recommended to modify temperature and top_p for the same completions request as the
            interaction of these two settings is difficult to predict.
            Supported range is [0, 1].
          default: 1
        max_tokens:
          type: integer
          format: int32
          minimum: 0
          description: The maximum number of tokens to generate.
        response_format:
          allOf:
            - $ref: '#/components/schemas/ChatCompletionsResponseFormat'
          description: |-
            An object specifying the format that the model must output.

            Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured Outputs which ensures the model will match your supplied JSON schema.

            Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the message the model generates is valid JSON.

            **Important:** when using JSON mode, you **must** also instruct the model to produce JSON yourself via a system or user message. Without this, the model may generate an unending stream of whitespace until the generation reaches the token limit, resulting in a long-running and seemingly "stuck" request. Also note that the message content may be partially cut off if `finish_reason="length"`, which indicates the generation exceeded `max_tokens` or the conversation exceeded the max context length.
        stop:
          type: array
          items:
            type: string
          minItems: 1
          description: A collection of textual sequences that will end completions generation.
        tools:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionsToolDefinition'
          minItems: 1
          description: |-
            A list of tools the model may request to call. Currently, only functions are supported as a tool. The model
            may response with a function call request and provide the input arguments in JSON format for that function.
        tool_choice:
          anyOf:
            - $ref: '#/components/schemas/ChatCompletionsToolChoicePreset'
            - $ref: '#/components/schemas/ChatCompletionsNamedToolChoice'
          description: If specified, the model will configure which of the provided tools it can use for the chat completions response.
        seed:
          type: integer
          format: int64
          description: |-
            If specified, the system will make a best effort to sample deterministically such that repeated requests with the
            same seed and parameters should return the same result. Determinism is not guaranteed.
        model:
          type: string
          description: ID of the specific AI model to use, if more than one model is available on the endpoint.
        modalities:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionsModality'
          description: |-
            The modalities that the model is allowed to use for the chat completions response. The default modality
            is `text`. Indicating an unsupported modality combination results in an 422 error.
      additionalProperties: {}
      description: |-
        The configuration information for a chat completions request.
        Completions support a wide variety of tasks and generate text that continues from or "completes"
        provided prompt data.
    ChatCompletionsResponseFormat:
      type: object
      required:
        - type
      properties:
        type:
          type: string
          description: The response format type to use for chat completions.
      discriminator:
        propertyName: type
        mapping:
          text: '#/components/schemas/ChatCompletionsResponseFormatText'
          json_object: '#/components/schemas/ChatCompletionsResponseFormatJsonObject'
          json_schema: '#/components/schemas/ChatCompletionsResponseFormatJsonSchema'
      description: |-
        Represents the format that the model must output. Use this to enable JSON mode instead of the default text mode.
        Note that to enable JSON mode, some AI models may also require you to instruct the model to produce JSON
        via a system or user message.
    ChatCompletionsResponseFormatJsonObject:
      type: object
      required:
        - type
      properties:
        type:
          type: string
          enum:
            - json_object
          description: "Response format type: always 'json_object' for this object."
      allOf:
        - $ref: '#/components/schemas/ChatCompletionsResponseFormat'
      description: |-
        A response format for Chat Completions that restricts responses to emitting valid JSON objects.
        Note that to enable JSON mode, some AI models may also require you to instruct the model to produce JSON
        via a system or user message.
    ChatCompletionsResponseFormatJsonSchema:
      type: object
      required:
        - type
        - json_schema
      properties:
        type:
          type: string
          enum:
            - json_schema
          description: 'The type of response format being defined: `json_schema`'
        json_schema:
          allOf:
            - $ref: '#/components/schemas/ChatCompletionsResponseFormatJsonSchemaDefinition'
          description: The definition of the required JSON schema in the response, and associated metadata.
      allOf:
        - $ref: '#/components/schemas/ChatCompletionsResponseFormat'
      description: |-
        A response format for Chat Completions that restricts responses to emitting valid JSON objects, with a
        JSON schema specified by the caller.
    ChatCompletionsResponseFormatJsonSchemaDefinition:
      type: object
      required:
        - name
        - schema
      properties:
        name:
          type: string
          description: The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.
        schema:
          type: object
          additionalProperties: {}
          description: The definition of the JSON schema
        description:
          type: string
          description: A description of the response format, used by the AI model to determine how to generate responses in this format.
        strict:
          type: boolean
          description: |-
            Whether to enable strict schema adherence when generating the output.
            If set to true, the model will always follow the exact schema defined in the `schema` field. Only a subset of
            JSON Schema is supported when `strict` is `true`.
          default: false
      description: The definition of the required JSON schema in the response, and associated metadata.
    ChatCompletionsResponseFormatText:
      type: object
      required:
        - type
      properties:
        type:
          type: string
          enum:
            - text
          description: "Response format type: always 'text' for this object."
      allOf:
        - $ref: '#/components/schemas/ChatCompletionsResponseFormat'
      description: A response format for Chat Completions that emits text responses. This is the default response format.
    ChatCompletionsToolCall:
      type: object
      required:
        - id
        - type
        - function
      properties:
        id:
          type: string
          description: The ID of the tool call.
        type:
          type: string
          enum:
            - function
          description: The type of tool call. Currently, only `function` is supported.
        function:
          allOf:
            - $ref: '#/components/schemas/FunctionCall'
          description: The details of the function call requested by the AI model.
      description: A function tool call requested by the AI model.
    ChatCompletionsToolChoicePreset:
      anyOf:
        - type: string
        - type: string
          enum:
            - auto
            - none
            - required
      description: Represents a generic policy for how a chat completions tool may be selected.
    ChatCompletionsToolDefinition:
      type: object
      required:
        - type
        - function
      properties:
        type:
          type: string
          enum:
            - function
          description: The type of the tool. Currently, only `function` is supported.
        function:
          allOf:
            - $ref: '#/components/schemas/FunctionDefinition'
          description: The function definition details for the function tool.
      description: The definition of a chat completions tool that can call a function.
    ChatMessageAudioContentItem:
      type: object
      required:
        - type
        - audio_url
      properties:
        type:
          type: string
          enum:
            - audio_url
          description: "The discriminated object type: always 'image_url' for this type."
        audio_url:
          allOf:
            - $ref: '#/components/schemas/ChatMessageAudioUrl'
          description: An internet location, which must be accessible to the model, from which the audio may be retrieved.
      allOf:
        - $ref: '#/components/schemas/ChatMessageContentItem'
      description: A structured chat content item containing an audio reference.
    ChatMessageAudioUrl:
      type: object
      required:
        - url
      properties:
        url:
          type: string
          description: The URL of the audio.
      description: An internet location from which the model may retrieve an audio.
    ChatMessageContentItem:
      type: object
      required:
        - type
      properties:
        type:
          type: string
          description: The discriminated object type.
      discriminator:
        propertyName: type
        mapping:
          text: '#/components/schemas/ChatMessageTextContentItem'
          image_url: '#/components/schemas/ChatMessageImageContentItem'
          audio_url: '#/components/schemas/ChatMessageAudioContentItem'
          input_audio: '#/components/schemas/ChatMessageInputAudioContentItem'
      description: An abstract representation of a structured content item within a chat message.
    ChatMessageImageContentItem:
      type: object
      required:
        - type
        - image_url
      properties:
        type:
          type: string
          enum:
            - image_url
          description: "The discriminated object type: always 'image_url' for this type."
        image_url:
          allOf:
            - $ref: '#/components/schemas/ChatMessageImageUrl'
          description: An internet location, which must be accessible to the model,from which the image may be retrieved.
      allOf:
        - $ref: '#/components/schemas/ChatMessageContentItem'
      description: A structured chat content item containing an image reference.
    ChatMessageImageDetailLevel:
      anyOf:
        - type: string
        - type: string
          enum:
            - auto
            - low
            - high
      description: A representation of the possible image detail levels for image-based chat completions message content.
    ChatMessageImageUrl:
      type: object
      required:
        - url
      properties:
        url:
          type: string
          description: The URL of the image.
        detail:
          allOf:
            - $ref: '#/components/schemas/ChatMessageImageDetailLevel'
          description: |-
            The evaluation quality setting to use, which controls relative prioritization of speed, token consumption, and
            accuracy.
      description: An internet location from which the model may retrieve an image.
    ChatMessageInputAudio:
      type: object
      required:
        - data
        - format
      properties:
        data:
          type: string
          description: Base64 encoded audio data
        format:
          allOf:
            - $ref: '#/components/schemas/AudioContentFormat'
          description: The audio format of the audio content.
      description: The details of an audio chat message content part.
    ChatMessageInputAudioContentItem:
      type: object
      required:
        - type
        - format
      properties:
        type:
          type: string
          enum:
            - input_audio
          description: "The discriminated object type: always 'input_audio' for this type."
        format:
          allOf:
            - $ref: '#/components/schemas/AudioContentFormat'
          description: The audio format of the audio reference.
      allOf:
        - $ref: '#/components/schemas/ChatMessageContentItem'
      description: A structured chat content item containing an audio content.
    ChatMessageTextContentItem:
      type: object
      required:
        - type
        - text
      properties:
        type:
          type: string
          enum:
            - text
          description: "The discriminated object type: always 'text' for this type."
        text:
          type: string
          description: The content of the message.
      allOf:
        - $ref: '#/components/schemas/ChatMessageContentItem'
      description: A structured chat content item containing plain text.
    ChatRequestAssistantMessage:
      type: object
      required:
        - role
      properties:
        role:
          type: string
          enum:
            - assistant
          description: The chat role associated with this message, which is always 'assistant' for assistant messages.
        content:
          type: string
          description: The content of the message.
        tool_calls:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionsToolCall'
          description: |-
            The tool calls that must be resolved and have their outputs appended to subsequent input messages for the chat
            completions request to resolve as configured.
        audio:
          allOf:
            - $ref: '#/components/schemas/ChatRequestAudioReference'
          description: '  The audio generated by a previous response in a multi-turn conversation.'
      allOf:
        - $ref: '#/components/schemas/ChatRequestMessage'
      description: A request chat message representing response or action from the assistant.
    ChatRequestAudioReference:
      type: object
      required:
        - id
      properties:
        id:
          type: string
          description: '  Unique identifier for the audio response. This value corresponds to the id of a previous audio completion.'
      description: A reference to an audio response generated by the model.
    ChatRequestMessage:
      type: object
      required:
        - role
      properties:
        role:
          allOf:
            - $ref: '#/components/schemas/ChatRole'
          description: The chat role associated with this message.
      discriminator:
        propertyName: role
        mapping:
          system: '#/components/schemas/ChatRequestSystemMessage'
          user: '#/components/schemas/ChatRequestUserMessage'
          assistant: '#/components/schemas/ChatRequestAssistantMessage'
          tool: '#/components/schemas/ChatRequestToolMessage'
      description: An abstract representation of a chat message as provided in a request.
    ChatRequestSystemMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - system
          description: The chat role associated with this message, which is always 'system' for system messages.
        content:
          type: string
          description: The contents of the system message.
      allOf:
        - $ref: '#/components/schemas/ChatRequestMessage'
      description: |-
        A request chat message containing system instructions that influence how the model will generate a chat completions
        response.
    ChatRequestToolMessage:
      type: object
      required:
        - role
        - tool_call_id
      properties:
        role:
          type: string
          enum:
            - tool
          description: The chat role associated with this message, which is always 'tool' for tool messages.
        content:
          type: string
          description: The content of the message.
        tool_call_id:
          type: string
          description: The ID of the tool call resolved by the provided content.
      allOf:
        - $ref: '#/components/schemas/ChatRequestMessage'
      description: A request chat message representing requested output from a configured tool.
    ChatRequestUserMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - user
          description: The chat role associated with this message, which is always 'user' for user messages.
        content:
          anyOf:
            - type: string
            - type: array
              items:
                $ref: '#/components/schemas/ChatMessageContentItem'
          description: The contents of the user message, with available input types varying by selected model.
      allOf:
        - $ref: '#/components/schemas/ChatRequestMessage'
      description: A request chat message representing user input to the assistant.
    ChatResponseMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          allOf:
            - $ref: '#/components/schemas/ChatRole'
          description: The chat role associated with the message.
          readOnly: true
        content:
          type: string
          nullable: true
          description: The content of the message.
          readOnly: true
        reasoning_content:
          type: string
          description: The reasoning content the model used for generating the response
          readOnly: true
        tool_calls:
          type: array
          items:
            $ref: '#/components/schemas/ChatCompletionsToolCall'
          description: |-
            The tool calls that must be resolved and have their outputs appended to subsequent input messages for the chat
            completions request to resolve as configured.
          readOnly: true
        audio:
          allOf:
            - $ref: '#/components/schemas/ChatCompletionsAudio'
          description: '  The audio generated by the model as a response to the messages if the model is configured to generate audio.'
          readOnly: true
      description: A representation of a chat message as received in a response.
    ChatRole:
      anyOf:
        - type: string
        - type: string
          enum:
            - system
            - developer
            - user
            - assistant
            - tool
      description: A description of the intended purpose of a message within a chat completions interaction.
    CompletionsFinishReason:
      anyOf:
        - type: string
        - type: string
          enum:
            - stop
            - length
            - content_filter
            - tool_calls
      description: Representation of the manner in which a completions response concluded.
    CompletionsUsage:
      type: object
      required:
        - completion_tokens
        - prompt_tokens
        - total_tokens
      properties:
        completion_tokens:
          type: integer
          format: int32
          description: The number of tokens generated across all completions emissions.
          readOnly: true
        prompt_tokens:
          type: integer
          format: int32
          description: The number of tokens in the provided prompts for the completions request.
          readOnly: true
        total_tokens:
          type: integer
          format: int32
          description: The total number of tokens processed for the completions request and response.
          readOnly: true
        completion_tokens_details:
          allOf:
            - $ref: '#/components/schemas/CompletionsUsageDetails'
          description: Breakdown of tokens used in a completion.
          readOnly: true
        prompt_tokens_details:
          allOf:
            - $ref: '#/components/schemas/PromptUsageDetails'
          description: Breakdown of tokens used in the prompt/chat history.
          readOnly: true
      description: |-
        Representation of the token counts processed for a completions request.
        Counts consider all tokens across prompts, choices, choice alternates, best_of generations, and
        other consumers.
    CompletionsUsageDetails:
      type: object
      required:
        - audio_tokens
        - reasoning_tokens
        - total_tokens
      properties:
        audio_tokens:
          type: integer
          format: int32
          description: The number of tokens corresponding to audio input.
          readOnly: true
        reasoning_tokens:
          type: integer
          format: int32
          description: The number of tokens corresponding to reasoning.
          readOnly: true
        total_tokens:
          type: integer
          format: int32
          description: The total number of tokens processed for the completions request and response.
          readOnly: true
      description: A breakdown of tokens used in a completion.
    EmbeddingEncodingFormat:
      anyOf:
        - type: string
        - type: string
          enum:
            - base64
            - binary
            - float
            - int8
            - ubinary
            - uint8
      description: |-
        Specifies the types of embeddings to generate. Compressed embeddings types like `uint8`, `int8`, `ubinary` and 
        `binary`, may reduce storage costs without sacrificing the integrity of the data. Returns a 422 error if the
        model doesn't support the value or parameter. Read the model's documentation to know the values supported by
        the your model.
    EmbeddingInputType:
      anyOf:
        - type: string
        - type: string
          enum:
            - text
            - query
            - document
      description: Represents the input types used for embedding search.
    EmbeddingItem:
      type: object
      required:
        - embedding
        - index
        - object
      properties:
        embedding:
          type: array
          items:
            type: number
            format: float
          description: |-
            List of embedding values for the input prompt. These represent a measurement of the
            vector-based relatedness of the provided input. Or a base64 encoded string of the embedding vector.
          readOnly: true
        index:
          type: integer
          format: int32
          description: Index of the prompt to which the EmbeddingItem corresponds.
          readOnly: true
        object:
          type: string
          enum:
            - embedding
          description: The object type of this embeddings item. Will always be `embedding`.
      description: Representation of a single embeddings relatedness comparison.
    EmbeddingsOptions:
      type: object
      required:
        - input
      properties:
        input:
          type: array
          items:
            type: string
          description: |-
            Input text to embed, encoded as a string or array of tokens.
            To embed multiple inputs in a single request, pass an array
            of strings or array of token arrays.
        dimensions:
          type: integer
          format: int32
          description: |-
            Optional. The number of dimensions the resulting output embeddings should have.
            Passing null causes the model to use its default value.
            Returns a 422 error if the model doesn't support the value or parameter.
        encoding_format:
          allOf:
            - $ref: '#/components/schemas/EmbeddingEncodingFormat'
          description: Optional. The desired format for the returned embeddings.
        input_type:
          allOf:
            - $ref: '#/components/schemas/EmbeddingInputType'
          description: |-
            Optional. The type of the input.
            Returns a 422 error if the model doesn't support the value or parameter.
        model:
          type: string
          description: ID of the specific AI model to use, if more than one model is available on the endpoint.
      additionalProperties: {}
      description: The configuration information for an embeddings request.
    EmbeddingsResult:
      type: object
      required:
        - id
        - data
        - usage
        - object
        - model
      properties:
        id:
          type: string
          description: Unique identifier for the embeddings result.
          readOnly: true
        data:
          type: array
          items:
            $ref: '#/components/schemas/EmbeddingItem'
          description: Embedding values for the prompts submitted in the request.
          readOnly: true
        usage:
          allOf:
            - $ref: '#/components/schemas/EmbeddingsUsage'
          description: Usage counts for tokens input using the embeddings API.
          readOnly: true
        object:
          type: string
          enum:
            - list
          description: The object type of the embeddings result. Will always be `list`.
        model:
          type: string
          description: The model ID used to generate this result.
          readOnly: true
      description: |-
        Representation of the response data from an embeddings request.
        Embeddings measure the relatedness of text strings and are commonly used for search, clustering,
        recommendations, and other similar scenarios.
    EmbeddingsUsage:
      type: object
      required:
        - prompt_tokens
        - total_tokens
      properties:
        prompt_tokens:
          type: integer
          format: int32
          description: Number of tokens in the request.
          readOnly: true
        total_tokens:
          type: integer
          format: int32
          description: |-
            Total number of tokens transacted in this request/response. Should equal the
            number of tokens in the request.
          readOnly: true
      description: Measurement of the amount of tokens used in this request and response.
    ExtraParameters:
      anyOf:
        - type: string
        - type: string
          enum:
            - error
            - drop
            - pass-through
      description: Controls what happens if extra parameters, undefined by the REST API, are passed in the JSON request payload.
    FunctionCall:
      type: object
      required:
        - name
        - arguments
      properties:
        name:
          type: string
          description: The name of the function to call.
          readOnly: true
        arguments:
          type: string
          description: |-
            The arguments to call the function with, as generated by the model in JSON format.
            Note that the model does not always generate valid JSON, and may hallucinate parameters
            not defined by your function schema. Validate the arguments in your code before calling
            your function.
          readOnly: true
      description: The name and arguments of a function that should be called, as generated by the model.
    FunctionDefinition:
      type: object
      required:
        - name
      properties:
        name:
          type: string
          description: The name of the function to be called.
        description:
          type: string
          description: |-
            A description of what the function does. The model will use this description when selecting the function and
            interpreting its parameters.
        parameters:
          type: object
          additionalProperties: {}
          description: The parameters the function accepts, described as a JSON Schema object.
      description: The definition of a caller-specified function that chat completions may invoke in response to matching user input.
    ImageEmbeddingInput:
      type: object
      required:
        - image
      properties:
        image:
          type: string
          description: 'The input image encoded in base64 string as a data URL. Example: `data:image/{format};base64,{data}`.'
        text:
          type: string
          description: |-
            Optional. The text input to feed into the model (like DINO, CLIP).
            Returns a 422 error if the model doesn't support the value or parameter.
      description: Represents an image with optional text.
    ImageEmbeddingsOptions:
      type: object
      required:
        - input
      properties:
        input:
          type: array
          items:
            $ref: '#/components/schemas/ImageEmbeddingInput'
          description: |-
            Input image to embed. To embed multiple inputs in a single request, pass an array.
            The input must not exceed the max input tokens for the model.
        dimensions:
          type: integer
          format: int32
          description: |-
            Optional. The number of dimensions the resulting output embeddings should have.
            Passing null causes the model to use its default value.
            Returns a 422 error if the model doesn't support the value or parameter.
        encoding_format:
          allOf:
            - $ref: '#/components/schemas/EmbeddingEncodingFormat'
          description: |-
            Optional. The number of dimensions the resulting output embeddings should have.
            Passing null causes the model to use its default value.
            Returns a 422 error if the model doesn't support the value or parameter.
        input_type:
          allOf:
            - $ref: '#/components/schemas/EmbeddingInputType'
          description: |-
            Optional. The type of the input.
            Returns a 422 error if the model doesn't support the value or parameter.
        model:
          type: string
          description: ID of the specific AI model to use, if more than one model is available on the endpoint.
      additionalProperties: {}
      description: The configuration information for an image embeddings request.
    ModelInfo:
      type: object
      required:
        - model_name
        - model_type
        - model_provider_name
      properties:
        model_name:
          type: string
          description: 'The name of the AI model. For example: `Phi21`'
          readOnly: true
        model_type:
          allOf:
            - $ref: '#/components/schemas/ModelType'
          description: The type of the AI model. A Unique identifier for the profile.
          readOnly: true
        model_provider_name:
          type: string
          description: 'The model provider name. For example: `Microsoft`'
          readOnly: true
      description: Represents some basic information about the AI model.
    ModelType:
      anyOf:
        - type: string
        - type: string
          enum:
            - embeddings
            - chat-completion
      description: The type of AI model
    PromptUsageDetails:
      type: object
      required:
        - audio_tokens
        - cached_tokens
      properties:
        audio_tokens:
          type: integer
          format: int32
          description: The number of tokens corresponding to audio input.
          readOnly: true
        cached_tokens:
          type: integer
          format: int32
          description: The total number of tokens cached.
          readOnly: true
      description: A breakdown of tokens used in the prompt/chat history.
    StreamingChatChoiceUpdate:
      type: object
      required:
        - index
        - finish_reason
        - delta
      properties:
        index:
          type: integer
          format: int32
          description: The ordered index associated with this chat completions choice.
        finish_reason:
          allOf:
            - $ref: '#/components/schemas/CompletionsFinishReason'
          nullable: true
          description: The reason that this chat completions choice completed its generated.
          readOnly: true
        delta:
          allOf:
            - $ref: '#/components/schemas/StreamingChatResponseMessageUpdate'
          description: An update to the chat message for a given chat completions prompt.
          readOnly: true
      description: |-
        Represents an update to a single prompt completion when the service is streaming updates 
        using Server Sent Events (SSE).
        Generally, `n` choices are generated per provided prompt with a default value of 1.
        Token limits and other settings may limit the number of choices generated.
    StreamingChatCompletionsUpdate:
      type: object
      required:
        - id
        - object
        - created
        - model
        - choices
      properties:
        id:
          type: string
          description: A unique identifier associated with this chat completions response.
        object:
          type: string
          enum:
            - chat.completion
          description: The response object type, which is always `chat.completion`.
        created:
          type: integer
          format: unixtime
          description: |-
            The first timestamp associated with generation activity for this completions response,
            represented as seconds since the beginning of the Unix epoch of 00:00 on 1 Jan 1970.
          readOnly: true
        model:
          type: string
          description: The model used for the chat completion.
          readOnly: true
        choices:
          type: array
          items:
            $ref: '#/components/schemas/StreamingChatChoiceUpdate'
          minItems: 1
          description: |-
            An update to the collection of completion choices associated with this completions response.
            Generally, `n` choices are generated per provided prompt with a default value of 1.
            Token limits and other settings may limit the number of choices generated.
          readOnly: true
        usage:
          allOf:
            - $ref: '#/components/schemas/CompletionsUsage'
          description: Usage information for tokens processed and generated as part of this completions operation.
          readOnly: true
      description: |-
        Represents a response update to a chat completions request, when the service is streaming updates 
        using Server Sent Events (SSE).
        Completions support a wide variety of tasks and generate text that continues from or "completes"
        provided prompt data.
    StreamingChatResponseMessageUpdate:
      type: object
      properties:
        role:
          allOf:
            - $ref: '#/components/schemas/ChatRole'
          description: The chat role associated with the message. If present, should always be 'assistant'
          readOnly: true
        content:
          type: string
          description: The content of the message.
          readOnly: true
        reasoning_content:
          type: string
          description: The reasoning content the model used for generating the response
          readOnly: true
        tool_calls:
          type: array
          items:
            $ref: '#/components/schemas/StreamingChatResponseToolCallUpdate'
          description: |-
            The tool calls that must be resolved and have their outputs appended to subsequent input messages for the chat
            completions request to resolve as configured.
          readOnly: true
      description: A representation of a chat message update as received in a streaming response.
    StreamingChatResponseToolCallUpdate:
      type: object
      required:
        - id
        - function
      properties:
        id:
          type: string
          description: The ID of the tool call.
          readOnly: true
        function:
          allOf:
            - $ref: '#/components/schemas/FunctionCall'
          description: Updates to the function call requested by the AI model.
          readOnly: true
      description: An update to the function tool call information requested by the AI model.
    Versions:
      type: string
      enum:
        - 2024-05-01-preview
        - '2025-04-01'
        - '2025-05-01'
        - 2025-05-15-preview
      description: The AI.Model service versions.
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: api-key
    BearerAuth:
      type: http
      scheme: Bearer
    OAuth2Auth:
      type: oauth2
      flows:
        implicit:
          authorizationUrl: https://login.microsoftonline.com/common/oauth2/v2.0/authorize
          scopes:
            https://cognitiveservices.azure.com/.default: ''
servers:
  - url: https://{resource}.services.ai.azure.com/api/models
    description: AI Model Inference
    variables:
      resource:
        default: ''
        description: The Azure AI Services resource name, for example 'my-resource'