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.
FreeOpen source
What you get
Youssofal MTPLX is a free, open-source AI tool published on GitHub by youssofal.
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.
GitHub stars: 2,625
License: Apache-2.0 (free to use)
Language: Python
Topics: anthropic-compatible, apple-silicon, claude-code, flash-next, inference-engine, llm-inference, local-ai, local-llm
Last updated: 2026-10-10
How to get it: open the repository on GitHub, read the README and follow the setup steps. HEEEXY lists it in the Lab because it is popular, actively maintained and openly licensed.
More for local ai
- Local Deep Research: ~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
- MakazhanAlpamys Soup: Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
- Dograh Hq Dograh: Open source voice AI platform. Self-hosted alternative to Vapi and Retell. On Prem, BYOK across Speech to Speech or LLM/STT/TTS, with a visual workflow builder, MCP native and telephony support.