UnifyRoute is a self-hosted, OpenAI-compatible gateway for routing requests across multiple LLM providers with failover, quota awareness, and a management UI.
- OpenAI-compatible API endpoints for easy SDK/tool integration.
- Tier-based routing with fallback behavior and provider redundancy.
- Credential and provider management through an interactive dashboard.
- Cost, usage, and operational visibility for production workflows.
SELF-HOSTEDOPENAI-COMPATIBLEMULTI-PROVIDER-ROUTINGFAILOVER-READYOPS-FRIENDLY
Linux/macOS:
git clone https://github.com/unifyroute/UnifyRoute.git
cd UnifyRoute
cp sample.env .env
./unifyroute setup
./unifyroute startWindows (Command Prompt):
git clone https://github.com/unifyroute/UnifyRoute.git
cd UnifyRoute
copy sample.env .env
unifyroute.bat setup
unifyroute.bat startDashboard: http://localhost:6565
👉 Windows users: Use unifyroute.bat instead of ./unifyroute. See Windows Setup Guide for detailed instructions and troubleshooting.
- Windows Setup Guide - For Windows users — Complete guide with troubleshooting
- Setup Guide (Cross-Platform) - Complete setup instructions for Windows, macOS, and Linux
- Getting Started
- CLI Reference
- Configuration Reference
- Architecture Overview
- Development Guide
- Migration Guide
api-gateway/ FastAPI gateway API and routes
router/ Routing engine and provider adapters
shared/ Shared models, schemas, and security helpers
gui/ React dashboard
credential-vault/ OAuth/secret helper service
quota-poller/ Quota and model sync workers
launcher/ Unified app launcher
selfheal/ Proactive failover and incident tracking
docs/ Documentation
scripts/ Utility and setup scripts
UnifyRoute exposes OpenAI-style endpoints:
POST /api/v1/chat/completionsPOST /api/v1/completionsGET /api/v1/models
Apache License 2.0. See LICENSE.