def wait_for_vllm_to_start()

in ultravox/inference/run_vllm_inference.py [0:0]

• 12 lines of code
• 4 McCabe index (conditional complexity)

def wait_for_vllm_to_start(port: int = 8000):
    while True:
        try:
            response = requests.get(f"http://localhost:{port}/health")
            response.raise_for_status()
            break
        except requests.exceptions.ConnectionError:
            pass
        except requests.exceptions.HTTPError:
            pass

        print("Waiting for server to start...")
        time.sleep(2)