An open source LLM gateway that routes one OpenAI-compatible key to models from every major AI provider, with failover.
Most teams that build with AI end up with a key and a bill at every provider they test. UnoRouter replaces that with one OpenAI-compatible endpoint: you keep your existing SDK, change the base URL, and pick GPT, Claude, Gemini, DeepSeek, Qwen or an open model by name. When an upstream errors or stalls, the call is retried on another provider, so a vendor incident does not turn into downtime for your users. It works with Claude Code, Cline, OpenCode, SillyTavern and RisuAI. The code is open source, the built-in chat stores conversations in your browser, and billing is a free tier followed by pay as you go credits without a subscription.
No reviews yet. Be the first to review UnoRouter.