Built on the shoulders of giants:
| Project | What It Does | By |
|---|---|---|
| 🤖 Claude Code | AI coding agent | Anthropic |
| 🍎 MLX + mlx-lm | Apple Silicon ML framework + inference | Apple |
| 🟢 Gemma 4 | Gemma 4 31B, 12B and E4B (base weights) | Google DeepMind |
| 🟣 Qwen 3.8 | Qwen 3.8 27B | Alibaba Qwen team |
| 🟡 Hermes 4 | Hermes 4 14B | Nous Research |
| ✨ Muse Glimmer | Muse Glimmer 30B | Meta |
| 👁️ Nemotron 3 Nano Omni | Nemotron Omni | NVIDIA |
| ⚡ DFlash 2 + mlx-dspark | Speculative decoding drafter and engine for Qwen 3.8 | Inco AI / Z Lab, ARahim3 |
| 📦 mlx-community + lmstudio-community | MLX conversions we install | — |
| 🔧 huihui-ai, null-space, OpenYourMind + Babsie | Abliterations we build on | — |
| 📖 Abliteration explained | The technique | Maxime Labonne |
The people behind the two libraries everything here runs on, by commit count:
- MLX: Awni Hannun, Cheng, Angelos Katharopoulos, Ayaan Gazali, Jagrit Digani and everyone else who has contributed.
- mlx-lm: Awni Hannun, Angelos Katharopoulos, Gökdeniz Gülmez, Anchen and everyone else who has contributed.
Contributors to this repo are credited in the README and the full guide.