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.
- 2.6k stars
- 206 forks
- 98 open issues
- Language: Python
- License: Apache-2.0
- updated
Install
git clone https://github.com/youssofal/MTPLX- Clone the repository.
- Follow the README for installation and usage instructions.
Tools overview
What people ask about MTPLX
What is youssofal/MTPLX?
youssofal/MTPLX is tools for the Claude AI ecosystem. 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. It has 2.6k GitHub stars and its last recorded update is dated 2026-10-10.
How do I install MTPLX?
You can install MTPLX by cloning the repository (https://github.com/youssofal/MTPLX) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is youssofal/MTPLX safe to use?
Our security agent has analyzed youssofal/MTPLX and assigned a Trust Score of 97/100 (tier: Verified). See the full breakdown of passed checks and flags on this page.
Who maintains youssofal/MTPLX?
youssofal/MTPLX is maintained by youssofal. The last recorded GitHub activity is dated 2026-10-10, with 98 open issues.
Are there alternatives to MTPLX?
Yes. On ClaudeWave you can browse similar tools at /categories/tools, sorted by popularity or recent activity.