MLX Community · First MLX build 3 Dec 2025
Ministral-3-14B-Instruct-2512
MLX
apache-2.0 Trending
A 14B parameter model, with 8 MLX builds, from MXFP4 to 5-bit.
- Parameters
- 14B From Hugging Face
- Quantisation
- 4-bit Ministral-3-14B-Instruct-2512-4bit
- Memory needed
- — GB Estimate, unified memory
- Download
- — Weights, estimated
- Downloads
- 3.4K This build, last 30 days
- Task
- —
- MLX builds
- 8 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/Ministral-3-14B-Instruct-2512-4bitMLX builds
8 on mlx-community| Ministral-3-14B-Instruct-2512-mxfp4 | MXFP4 | 8.4 GB | 12 GB | 83 |
|---|---|---|---|---|
| Ministral-3-14B-Instruct-2512-nvfp4 | 4-bit | 8.4 GB | 12 GB | 41 |
| Ministral-3-14B-Instruct-2512-mxfp8 | MXFP8 | 15.2 GB | 19 GB | 59 |
| Ministral-3-14B-Instruct-2512-4bit | 4-bit | — | — | 3.4K |
| Ministral-3-14B-Instruct-2512-8bit | 8-bit | — | — | 349 |
| Ministral-3-14B-Instruct-2512-6bit | 6-bit | — | — | 120 |
| Ministral-3-14B-Instruct-2512-bf16 | bf16 | — | — | 71 |
| Ministral-3-14B-Instruct-2512-5bit | 5-bit | — | — | 38 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.