MLX Community · First MLX build 19 Mar 2025
Mistral-Small-3.1-24B-Instruct-2503
MLX
apache-2.0
A 24B parameter model, with 5 MLX builds, from bf16 to 6-bit.
- Parameters
- 24B From Hugging Face
- Quantisation
- 4-bit Mistral-Small-3.1-24B-Instruct-2503-4bit
- Memory needed
- — GB Estimate, unified memory
- Download
- — Weights, estimated
- Downloads
- 1.5K This build, last 30 days
- Task
- —
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Hugging Face does not list a weights size for this build, so there is no memory estimate.
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-3.1-24B-Instruct-2503-4bitMLX builds
5 on mlx-community| Mistral-Small-3.1-24B-Instruct-2503-bf16 | bf16 | 48.0 GB | 56 GB | 58 |
|---|---|---|---|---|
| Mistral-Small-3.1-24B-Instruct-2503-4bit | 4-bit | — | — | 1.5K |
| Mistral-Small-3.1-24B-Instruct-2503-8bit | 8-bit | — | — | 147 |
| Mistral-Small-3.1-24B-Instruct-2503-3bit | 3-bit | — | — | 56 |
| Mistral-Small-3.1-24B-Instruct-2503-6bit | 6-bit | — | — | 51 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.