MTPLX is a native Mac app for local AI. Chat, agents, and a local server, all at twice the speed.
Twice the speed at no quality loss, at any temperature. The output is identical, verified bit for bit. How it works →
One click serves OpenCode, Pi, Hermes, or the web UI from your Mac. OpenAI and Anthropic compatible, so everything plugs in.
Native Swift, fully offline, beautiful. Watch your models write code and prose at twice the speed.
Paste a Hugging Face link. Forge converts it to MLX and measures the real speedup on your Mac.