About this project
This Node.js-based proxy layer allows users to route AI model requests across multiple providers using an OpenAI-compatible interface (/v1/chat/completions). It is designed to ensure reliability and flexibility in AI deployments.
Key capabilities include:
- Multi-provider fallback: Route requests to multiple providers for the same model based on defined priority.
- Health-aware routing: Tracks provider health in real-time to automate fallback and recovery processes.
- Secure configuration: Supports API key injection via environment variables to avoid hardcoding secrets.
- Model mapping: Uses a YAML configuration to map specific models to various provider endpoints and API keys.
- Local Admin UI: Includes an optional lightweight panel for managing model-provider mappings.
- Edge-ready design: Architected for potential migration to edge platforms like Cloudflare Workers or Vercel Edge.
Comments
0 Rating appears after 10 ratings
Sign in to join the discussion.