by youssofal · Claude Skill · ★ 2.7k
The fastest way to run Qwen 3.8 on a Mac. MTPLX is a native Mac app and a command line that runs local language models on Apple Silicon with the model's own multi-token prediction (MTP) heads. It runs Qwen 3.8 Flash Next, the 125B mixture of experts, and Qwen 3.8 27B, plus Qwen 3.6, Qwen 3.5 and Gemma 4. The model drafts several tokens ahead of itself, one batched forward pass verifies the draft, and tokens are committed through exact rejection sampling with residual correction.
| Stars | 2,741 |
| Forks | 215 |
| Language | Python |
| Category | Claude Skill |
| License | Apache-2.0 |
| Quality Score | 37.3/100 |
| Open Issues | 82 |
| Last Updated | 2026-11-22 |
| Created | 2026-05-02 |
| Platforms | claude-code, python |
| Est. Tokens | ~1710k |
These tools work well together with MTPLX for enhanced workflows:
Explore other popular claude skill tools:
MTPLX is The fastest way to run Inkling 2 Mini, Qwen 3.8 and Ternary Bonsai 2 27B on a Mac: Inkling 2 Mini at up to 128.8 tok/s and Flash Next at 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Ma. It is categorized as a Claude Skill with 2.7k GitHub stars.
MTPLX is primarily written in Python. It covers topics such as anthropic-compatible, apple-silicon, claude-code.
You can find installation instructions and usage details in the MTPLX GitHub repository at github.com/youssofal/MTPLX. The project has 2.7k stars and 215 forks, indicating an active community.
MTPLX is released under the Apache-2.0 license, making it free to use and modify according to the license terms.