How to beginner · 3 min read

How to deploy a model in Azure OpenAI

Quick answer

To deploy a model in Azure OpenAI, create a deployment in the Azure portal under your OpenAI resource, specifying the model and deployment name. Then use the AzureOpenAI Python client with your api_key and azure_endpoint to call the deployed model by its deployment name.

PREREQUISITES

Python 3.8+
Azure OpenAI resource with deployment created
Azure OpenAI API key
pip install openai>=1.0

Setup

Install the openai Python package version 1.0 or higher. Set environment variables for your Azure OpenAI API key, endpoint, and deployment name.

AZURE_OPENAI_API_KEY: Your Azure OpenAI API key
AZURE_OPENAI_ENDPOINT: Your Azure OpenAI endpoint URL (e.g., https://your-resource-name.openai.azure.com/)
AZURE_OPENAI_DEPLOYMENT: The deployment name you created in Azure portal

bash

pip install openai>=1.0

Step by step

This example shows how to call a deployed model in Azure OpenAI 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"
)

def main():
    response = client.chat.completions.create(
        model=os.environ["AZURE_OPENAI_DEPLOYMENT"],
        messages=[{"role": "user", "content": "Explain the benefits of Azure OpenAI."}]
    )
    print("Response:", response.choices[0].message.content)

if __name__ == "__main__":
    main()

output

Response: Azure OpenAI provides scalable, secure, and integrated AI services that enable developers to build intelligent applications with ease.

Common variations

You can use different models by creating separate deployments in the Azure portal and referencing their deployment names in model=. For streaming responses, use stream=True in chat.completions.create(). Async calls require an async client or wrapper since the official SDK is synchronous.

python

response = client.chat.completions.create(
    model=os.environ["AZURE_OPENAI_DEPLOYMENT"],
    messages=[{"role": "user", "content": "Tell me a joke."}],
    stream=True
)
for chunk in response:
    print(chunk.choices[0].delta.get('content', ''), end='')

output

Why did the AI go to school? To improve its neural network!

Troubleshooting

401 Unauthorized: Check your AZURE_OPENAI_API_KEY and ensure it has access to the resource.
404 Not Found: Verify your AZURE_OPENAI_ENDPOINT and deployment name are correct.
Model not found: Confirm the deployment exists in the Azure portal and is active.

✅

Key Takeaways

Create model deployments in the Azure portal before calling them via API.
Use the AzureOpenAI client with your endpoint, API key, and deployment name.
Streaming and multiple deployments are supported by changing parameters and deployment names.

Verified 2026-04 · gpt-4o, AzureOpenAI

Verify ↗