How to Intermediate · 3 min read

Azure OpenAI with Azure API Management

Q: Azure OpenAI with Azure API Management

Use AzureOpenAI client from the openai Python package configured with your Azure API Management endpoint and deployment name. Set azure_endpoint and api_key environment variables, then call chat.completions.create with your model deployment to securely access Azure OpenAI through API Management.

Quick answer

Use AzureOpenAI client from the openai Python package configured with your Azure API Management endpoint and deployment name. Set azure_endpoint and api_key environment variables, then call chat.completions.create with your model deployment to securely access Azure OpenAI through API Management.

PREREQUISITES

Python 3.8+
Azure OpenAI resource with deployed model
Azure API Management instance configured with OpenAI proxy
pip install openai>=1.0
Environment variables for Azure API key and endpoint

Setup

Install the official openai Python package and set environment variables for your Azure API Management endpoint and API key. Ensure your Azure OpenAI model is deployed and accessible via API Management.

bash

pip install openai>=1.0

output

Collecting openai
  Downloading openai-1.x.x-py3-none-any.whl (xx kB)
Installing collected packages: openai
Successfully installed openai-1.x.x

Step by step

This example demonstrates calling Azure OpenAI through Azure API Management using the AzureOpenAI client. Replace environment variables with your actual values.

python

import os
from openai import AzureOpenAI

client = AzureOpenAI(
    api_key=os.environ["AZURE_OPENAI_API_KEY"],
    azure_endpoint=os.environ["AZURE_OPENAI_ENDPOINT"],
    api_version="2024-02-01"
)

response = client.chat.completions.create(
    model=os.environ["AZURE_OPENAI_DEPLOYMENT"],
    messages=[{"role": "user", "content": "Hello from Azure API Management!"}]
)

print("Response:", response.choices[0].message.content)

output

Response: Hello from Azure API Management! How can I assist you today?

Common variations

Use async calls with asyncio and await client.chat.completions.acreate(...).
Stream responses by adding stream=True to chat.completions.create.
Switch models by changing the model parameter to another deployment name.

python

import asyncio
from openai import AzureOpenAI

async def main():
    client = AzureOpenAI(
        api_key=os.environ["AZURE_OPENAI_API_KEY"],
        azure_endpoint=os.environ["AZURE_OPENAI_ENDPOINT"],
        api_version="2024-02-01"
    )

    stream = await client.chat.completions.acreate(
        model=os.environ["AZURE_OPENAI_DEPLOYMENT"],
        messages=[{"role": "user", "content": "Stream this response."}],
        stream=True
    )

    async for chunk in stream:
        print(chunk.choices[0].delta.content or "", end="", flush=True)

asyncio.run(main())

output

Streamed response text appears here in real time...

Troubleshooting

If you get 401 Unauthorized, verify your AZURE_OPENAI_API_KEY and that API Management is correctly configured.
404 Not Found means the deployment name or endpoint URL is incorrect.
Timeouts may require checking network connectivity or increasing client timeout settings.

✅

Key Takeaways

Use AzureOpenAI client with azure_endpoint and api_key for API Management integration.
Set environment variables AZURE_OPENAI_API_KEY, AZURE_OPENAI_ENDPOINT, and AZURE_OPENAI_DEPLOYMENT for secure access.
Support async and streaming calls by using acreate and stream=True parameters.
Common errors relate to authentication and deployment naming; verify these first.
Azure API Management enables centralized control and security for Azure OpenAI usage.

Verified 2026-04 · gpt-4o, gpt-4o-mini, azure-openai-deployment

Verify ↗