MXFP4 Open on Hugging Face
Parameters
229B
Weights (est.)
≈ 129 GB
Mac memory (est.)
192 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/MiniMax-M2.1-mxfp4 --prompt "Hello"
Details
- Base model
- MiniMaxAI/MiniMax-M2.1 (quantized)
- Vendor
- MiniMaxAI
- Task
- Text generation
- Quantisation
- MXFP4
- Licence
- modified-mit
- Downloads
- 26 in the last 30 days · 854 all time · 0 likes
- Published
- 2026-02-11 · updated 2026-02-11 (9mo ago)
- Tags
- minimax_m2, custom_code
Other MLX builds of MiniMaxAI/MiniMax-M2.1
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| MiniMax-M2.1-3bit | 3-bit | ≈ 100 GB | 128 GB+ | 257 |
| MiniMax-M2.1-4bit | 4-bit | ≈ 129 GB | 192 GB+ | 937 |
| MiniMax-M2.1-mxfp4 (this one) | MXFP4 | ≈ 129 GB | 192 GB+ | 26 |
| MiniMax-M2.1-5bit | 5-bit | ≈ 157 GB | 192 GB+ | 34 |
| MiniMax-M2.1-6bit | 6-bit | ≈ 186 GB | 256 GB+ | 45 |
| MiniMax-M2.1-8bit | 8-bit | ≈ 243 GB | 512 GB+ | 154 |