GitHub · Python
youssofal/MTPLX
The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Macs. Native MTP speculative decoding on Apple Silicon, exact at any temperature. OpenAI and Anthropic compatible local server.
- Stars
- 2,730
- Forks
- 214
- Stars gained in 24 hours
- +4
- Forks gained in 24 hours
- 0
Topics: anthropic-compatible, apple-silicon, claude-code, flash-next, inference-engine, llm-inference, local-ai, local-llm, macos, metal
Latest data point: Nov 21, 2026 · 2,730 stars · 214 forks