Muon Gate

Bring your own AI on your terms. An LLM client that connects to OpenRouter or Groq with your own API key, or to a local Ollama endpoint for private, low-latency chat. No accounts, no lock‑in, no logs, and requests go directly to the endpoints you configure.


Features

• Connect your own provider
• OpenRouter: choose your preferred model and stream responses
• Groq: fast, low‑latency streaming with your API key
• Ollama: point to a local endpoint for fully local inference
• Private by design
• No sign‑in required
• Use your keys; control where requests are sent
• Works with local or self‑hosted endpoints
• Fast, streaming chat
• Real‑time token streaming with cancel/stop
• Reliable delivery status and error handling
• Thinking models have a thoughts modal
• Per‑provider model selection
• Custom HTTP/WebSocket endpoints for advanced setups
• Light/Dark mode support
• Privacy first, no chat logs and easy wipe of all chat history


Requirements

OpenRouter or Groq requires your own API key
Ollama requires a running endpoint (e.g., on your computer or LAN)


Notes

This app sends requests directly to the endpoints you configure. Keep your keys safe and only use providers you trust.
Not affiliated with OpenRouter, Groq, or Ollama.