MLX Community · First MLX build 18 Mar 2025
Mistral-Small-3.1-Text-24B-Instruct-2503
MLX
A 24B parameter model, with 5 MLX builds, from 3-bit to bf16.
- Parameters
- 24B From Hugging Face
- Quantisation
- 4-bit Mistral-Small-3.1-Text-24B-Instruct-2503-4bit
- Memory needed
- 6 GB Estimate, unified memory
- Download
- 3.1 GB Weights, estimated
- Downloads
- 38 This build, last 30 days
- Task
- —
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 6 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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/Mistral-Small-3.1-Text-24B-Instruct-2503-4bitMLX builds
5 on mlx-community| Mistral-Small-3.1-Text-24B-Instruct-2503-3bit | 3-bit | 2.4 GB | 5 GB | 18 |
|---|---|---|---|---|
| Mistral-Small-3.1-Text-24B-Instruct-2503-4bit | 4-bit | 3.1 GB | 6 GB | 38 |
| Mistral-Small-3.1-Text-24B-Instruct-2503-6bit | 6-bit | 5.1 GB | 8 GB | 17 |
| Mistral-Small-3.1-Text-24B-Instruct-2503-8bit | 8-bit | 7.7 GB | 11 GB | 21 |
| Mistral-Small-3.1-Text-24B-Instruct-2503-bf16 | bf16 | 47.1 GB | 55 GB | 16 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.