Chat with local AI models on your own hardware — no cloud, no subscriptions, no data leaving your device. Connect to any Ollama server and chat with your favorite open-source models (Llama, Qwen, Mistral, Gemma, and more). Voice input, text-to-speech, tool calling via MCP, document search with RAG — everything running on your machine, private by design. FEATURES • Connect to any Ollama instance (local, remote, or Tailscale) • Streaming chat with thinking model support • Photo attachments — ask...