Azure OpenAI Proxy

Azure OpenAI Proxy is an open-source adapter published on GitHub that converts official OpenAI API requests into Azure OpenAI API requests, so OpenAI-ecosystem tools can use Azure deployments unchanged. The project has gathered around 1,300 GitHub stars.

Overview

Azure OpenAI exposes the same models as OpenAI but with different endpoints, headers and deployment names, which breaks tools written against the OpenAI API. This small Go proxy sits in between and translates: point any OpenAI client at the proxy, configure your Azure endpoint, API version and a model mapper that links OpenAI model names to your Azure deployment names, and requests flow through — including GPT-4-class chat models, embeddings and streaming responses. It has been verified with popular clients such as LangChain, ChatGPT-Next-Web, Chatbox and chatgpt-web. Configuration comes from environment variables or a YAML file that can define different endpoints and keys per model, and the proxy ships as a single binary or Docker image, making it a five-minute fix for teams standardising on Azure. The code was last updated in 2024.

Key Features

• Translates OpenAI API calls to Azure OpenAI transparently
• Supports chat completions, embeddings and streaming
• Model mapper links OpenAI names to Azure deployment names
• Per-model endpoints and keys via a YAML config file
• Verified with LangChain, ChatGPT-Next-Web, Chatbox and more
• Single Go binary or Docker image deployment

💳 Pricing

Azure OpenAI Proxy is free and open source under the Apache 2.0 license. You only pay for your own Azure OpenAI usage.

Who Is It For

Developers and teams whose organisation uses Azure OpenAI but who want to keep using tools built for the OpenAI API.