Summary

This project provides an OpenAI-compatible proxy server for Freebuff, enabling users to leverage Freebuff's free LLM models through standard OpenAI API clients. It targets developers and power users seeking to integrate Freebuff's capabilities into existing workflows without direct Freebuff SDK integration, offering features like dynamic request handling and token rotation for increased throughput and stealth.

How It Works

Freebuff2API acts as an intermediary, translating incoming OpenAI API requests into Freebuff's native format. It employs dynamic, randomized client fingerprints to mimic official SDK behavior, enhancing stealth. The system supports rotating through multiple Freebuff authentication tokens, automatically cycling them at configurable intervals to manage rate limits and improve availability. Outbound traffic can be routed through an optional HTTP proxy.

Quick Start & Requirements

• Primary Install: Docker is recommended. A pre-built multi-arch image is available on GHCR. Example: docker run -d --name Freebuff2API -p 8080:8080 -e AUTH_TOKENS="token1,token2" ghcr.io/quorinex/freebuff2api:latest.
• Build from Source: Requires Go 1.23+. Clone the repository, build using go build, and run the executable.
• Prerequisites: One or more Freebuff authentication tokens are mandatory. These can be obtained from the Freebuff web interface (https://freebuff.llm.pm) or via the Freebuff CLI.
• Links: Freebuff web interface: https://freebuff.llm.pm.

Highlighted Details

• OpenAI Compatibility: Exposes standard OpenAI API endpoints, allowing seamless integration with any compatible client, SDK, or CLI tool.
• Stealth Request Handling: Implements dynamic, randomized client fingerprints to obscure the use of the proxy and mimic official Freebuff SDK behavior.
• Multi-Token Rotation: Supports cycling through multiple Freebuff auth tokens with automatic, periodic rotation for increased throughput and resilience.
• HTTP Proxy Support: Allows routing all outbound traffic through a configurable upstream HTTP proxy.

Maintenance & Community

The README does not mention specific contributors, community channels (like Discord/Slack), sponsorships, or a roadmap.

Licensing & Compatibility

• License: MIT.
• Compatibility: The MIT license generally permits commercial use and linking with closed-source projects.

Limitations & Caveats

The project is explicitly stated as not being production-ready and is provided on an "As-Is" basis. Users are warned that they use it at their own risk, and the author assumes no liability for damages. It is intended for communication, experimentation, and learning purposes.