Ad

cf-openai-azure-proxy: Proxy OpenAI requests to Azure OpenAI

This project provides a Cloudflare Worker to proxy requests from OpenAI-compatible clients to Azure OpenAI Service, enabling access to models like GPT-4 and DALL-E-3 without requiring a dedicated server.
Screenshot of haibbo/cf-openai-azure-proxy homepage

This project introduces cf-openai-azure-proxy, a Cloudflare Worker designed to bridge the gap between OpenAI's popular client libraries and the Azure OpenAI Service. It addresses the limitation where existing OpenAI clients do not natively support Azure OpenAI. The worker acts as a proxy, forwarding requests to Azure OpenAI and returning the responses to the client, allowing for seamless use of Azure's powerful models.

The project leverages the scalability and cost-effectiveness of Cloudflare Workers to provide a serverless proxy solution. It offers straightforward deployment via the Cloudflare dashboard and supports streaming responses for enhanced user experience. The design allows for easy addition of new models by modifying a simple mapping configuration.

  • Model Support: GPT-3, GPT-3.5-turbo, GPT-4, and DALL-E-3 are supported.
  • Serverless Architecture: Deploys as a Cloudflare Worker, eliminating the need for dedicated servers.
  • Streaming: Implements a printer-like mode for streaming responses, improving real-time interactions.
  • Easy Deployment: Simplifies deployment through the Cloudflare dashboard and environment variables.
  • Custom Domain Support: Allows for using custom domains for enhanced security and branding.

The project has been actively maintained, with recent commits indicating continued development and support. The documentation provides clear instructions for setup and configuration, and the community is responsive to reported issues. While relatively new, the project demonstrates solid architecture and practical functionality.

This project benefits developers and users who want to leverage the capabilities of Azure OpenAI Service with their existing OpenAI client code. It offers an accessible and cost-effective solution for accessing advanced models like GPT-4 and DALL-E-3 without the complexities of server management. It simplifies integration and unlocks the power of Azure OpenAI for a wider range of applications.

Summarize:
Share:
Stars
1,620
Forks
197
Issues
17
Created
3 years ago
Commit
2 years ago
License
MIT
Archived
No
Updated 10 days ago

Similar Repositories