Run many local models. One simple workspace.
Local-first privacy. Unified chat. OpenAI-compatible API. Zero-cloud by default.
Free and open source • No account required
How it works
1
Pick a model
Choose from local models via Ollama, llama.cpp, or MLX
2
Start chatting
Unified chat interface with streaming responses
3
Add files for context
Drag and drop files for local RAG capabilities
Designed for local-first workflows; familiar like LM Studio, flexible like AnythingLLM.
🚀 We're Live!
We're launching on Product Hunt!
Help us reach more developers who value local-first AI tools. Your support means everything to our community.
Takes just 30 seconds • Helps us reach more developers
Ready to get started?
Made for developers, researchers, and privacy-conscious teams.