MLX Community · First MLX build 11 Jul 2025
Devstral-Small-2507
MLX
Text apache-2.0
A 24B parameter text generation model, with 5 MLX builds, from 4-bit to 8-bit.
- Parameters
- 24B From Hugging Face
- Quantisation
- 4-bit Devstral-Small-2507-4bit
- Memory needed
- 17 GB Estimate, unified memory
- Download
- 13.3 GB Weights, estimated
- Downloads
- 213 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 17 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/Devstral-Small-2507-4bit --prompt "Hello"MLX builds
5 on mlx-community| Devstral-Small-2507-4bit | 4-bit | 13.3 GB | 17 GB | 213 |
|---|---|---|---|---|
| Devstral-Small-2507-4bit-DWQ | 4-bit DWQ | 13.3 GB | 17 GB | 183 |
| Devstral-Small-2507-5bit | 5-bit | 16.2 GB | 21 GB | 97 |
| Devstral-Small-2507-6bit | 6-bit | 19.2 GB | 24 GB | 82 |
| Devstral-Small-2507-8bit | 8-bit | 25.0 GB | 30 GB | 121 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.