MiniMax · First MLX build 26 Dec 2025
MiniMax-M2.1
MLX
Text modified-mit
A 229B parameter text generation model from MiniMax, with 6 MLX builds, from 3-bit to 8-bit.
- Parameters
- 229B From Hugging Face
- Quantisation
- 4-bit MiniMax-M2.1-4bit
- Memory needed
- 146 GB Estimate, unified memory
- Download
- 129 GB Weights, estimated
- Downloads
- 940 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 6 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 146 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/MiniMax-M2.1-4bit --prompt "Hello"MLX builds
6 on mlx-community| MiniMax-M2.1-3bit | 3-bit | 100 GB | 114 GB | 254 |
|---|---|---|---|---|
| MiniMax-M2.1-4bit | 4-bit | 129 GB | 146 GB | 940 |
| MiniMax-M2.1-mxfp4 | MXFP4 | 129 GB | 146 GB | 26 |
| MiniMax-M2.1-5bit | 5-bit | 157 GB | 178 GB | 34 |
| MiniMax-M2.1-6bit | 6-bit | 186 GB | 210 GB | 45 |
| MiniMax-M2.1-8bit | 8-bit | 243 GB | 274 GB | 146 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.