heartshare/openai-azure-proxy

A server to delegate OpenAI chat requests to Azure OpenAI service

openai-azure-proxy

A server to delegate OpenAI chat requests to Azure OpenAI service. Since now most ChatGPT clients did not support Azure OpenAI service.

Privacy Policy: I do not collect any data sent through this server. To avoid any accidential data leak, a self deployment is recommanded.

Goal:

• Make ChatGPT clients (OpenCat currently) work.
• Delegate Chat requests to Azure OpenAI service.

Non Goal:

• Support all OpenAI/ChatGPT APIs

How to use

Their is a deployed version on Google Cloud Run: https://openai-azure-proxy-wykw6ffdpa-de.a.run.app.

Setup Azure OpenAI

Refer to Azure OpenAI doc for details. Make sure gpt-35-turbo is deployed in your Azure OpenAI studio.

Settings in OpenCat

• Model: gpt-3.5-turbo
• API key: Your Azure OpenAI key
• Custom API Domain: openai-azure-proxy-wykw6ffdpa-de.a.run.app/azure/$YOUR ENDPOINT. Endpoint is in your Azure Portal, such as: https://your-deployment-id.openai.azure.com. Please use your-deployment-id part in the custom API domain. The full Custom API domain should be like: openai-azure-proxy-wykw6ffdpa-de.a.run.app/azure/YOUR_DEPLOYMENT_ID.
• Check Send API Key

[#] GPT4: Currently I do not have a GPT4 enabled Azure OpenAI deployment.

Self deployment

• Docker: https://github.com/richard1122/openai-azure-proxy/pkgs/container/openai-azure-proxy
• Spring Boot: Check source code to build with gradle.