llama-index-llms-azure-openai

What it is and what it does

This package bridges LlamaIndex applications to Azure OpenAI's managed language models. It wraps Azure's OpenAI API behind LlamaIndex's standard LLM interface, letting you use Azure-hosted GPT models (like gpt-35-turbo-16k) for text completion, streaming, and chat without rewriting your LlamaIndex code.

The package handles Azure-specific authentication through azure-identity, manages HTTP requests via httpx, and depends on llama-index-core and llama-index-llms-openai for the core LLM abstraction. You configure it with an Azure endpoint, API key, API version, and the name of your deployed model engine, then call standard methods like complete(), stream_complete(), chat(), and stream_chat() just as you would with any LlamaIndex LLM.

Use it for:

• Build RAG or agentic applications using Azure-hosted GPT models within LlamaIndex's framework
• Migrate existing OpenAI-based LlamaIndex code to run on Azure's managed infrastructure
• Stream long-form completions or multi-turn conversations through Azure OpenAI endpoints
• Integrate Azure OpenAI into enterprise applications that require Azure's compliance and security posture

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Integrates Azure OpenAI language models into LlamaIndex applications, enabling text completion, streaming, and chat interactions through Azure's managed OpenAI service.

Yes, if you are already using LlamaIndex and need to connect to Azure OpenAI models. The package is actively maintained, has no known vulnerabilities, and integrates cleanly with LlamaIndex's standard LLM interface. Install friction is low. You will need valid Azure OpenAI credentials and a deployed model to use it.

Install

llama-index-llms-azure-openai on PyPI

pip install llama-index-llms-azure-openai

uv add llama-index-llms-azure-openai

poetry add llama-index-llms-azure-openai

Installing llama-index-llms-azure-openai

Before you install

Low install friction with a pure-Python wheel. Maintenance status is active. Depends on llama-index-core and llama-index-llms-openai, plus azure-identity and httpx for Azure authentication and HTTP operations.

License in practice

MIT license permits commercial and private use with minimal restrictions, making it suitable for most projects.

Quickstart

import os
os.environ["OPENAI_API_KEY"] = "<your-api-key>"
os.environ["AZURE_OPENAI_ENDPOINT"] = "https://<your-resource-name>.openai.azure.com/"
os.environ["OPENAI_API_VERSION"] = "2023-07-01-preview"

from llama_index.llms.azure_openai import AzureOpenAI
llm = AzureOpenAI(engine="my-engine", model="gpt-35-turbo-16k", temperature=0.0)
response = llm.complete("The sky is a beautiful blue and")
print(response)

Requires an active Azure OpenAI resource with a deployed model and valid API credentials (endpoint, key, and API version).

Verify before relying

• Whether the package maintains backward compatibility across Azure OpenAI API versions beyond 2023-07-01-preview
• Performance characteristics when handling large-scale concurrent requests through Azure endpoints

Package facts

Evidence: llama_index_llms_azure_openai-0.5.5-py3-none-any.whl

Tags

More Artificial Intelligence packages