Turn your Mac into a managed local AI server. LLMs, vision, image generation, transcription, TTS, embeddings — controlled from the menu bar.
bash <(curl -fsSL raw.githubusercontent.com/jerrytalton/super-puppy/main/install.sh)
Start, stop, and monitor Ollama, MLX, and MCP services. Switch model profiles. See memory usage at a glance.
OpenAI-compatible endpoints for every model. Your existing code works without changes.
16 tools — generate, review, vision, image gen, transcription, TTS, translation, embeddings, and more — all running on your hardware.
Install on a desktop to serve models over Tailscale. Install on a laptop to auto-discover the desktop and fall back to local when away.
Test any capability from a browser. Works as a PWA on your phone — your mobile device becomes a frontend to your desktop GPU.
Tagged releases are checked every 2 minutes. Updates apply automatically with crash rollback. install.sh --uninstall for clean removal.