bf16 Open on Hugging Face
Parameters
8.9B
Weights (est.)
≈ 17.8 GB
Mac memory (est.)
24 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/Ministral-3-8B-Base-2512-bf16 --image photo.jpg --prompt "Describe this image"
Details
- Base model
- mistralai/Ministral-3-8B-Base-2512 (finetune)
- Vendor
- Mistral
- Task
- Vision + text
- Quantisation
- bf16
- Licence
- apache-2.0
- Downloads
- 53 in the last 30 days · 227 all time · 0 likes
- Published
- 2026-07-18 · updated 2026-07-18 (3mo ago)
- Languages
- en, fr, es, de, it, pt, nl, zh, ja, ko, ar
- Tags
- mistral3, mistral-common, ministral, ministral-3, vision-language, multimodal, quantized, edge, bf16, base-model
Other MLX builds of mistralai/Ministral-3-8B-Base-2512
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Ministral-3-8B-Base-2512-4bit | 4-bit | ≈ 4.8 GB | 8 GB+ | 107 |
| Ministral-3-8B-Base-2512-6bit | 6-bit | ≈ 8.4 GB | 16 GB+ | 39 |
| Ministral-3-8B-Base-2512-8bit | 8-bit | ≈ 10.4 GB | 16 GB+ | 53 |
| Ministral-3-8B-Base-2512-bf16 (this one) | bf16 | ≈ 17.8 GB | 24 GB+ | 53 |
| Ministral-3-8B-Reasoning-2512-bf16 | bf16 | ≈ 17.8 GB | 24 GB+ | 34 |
| Ministral-3-8B-Instruct-2512-4bit | 4-bit | — | — | 3.6K |
| Ministral-3-8B-Reasoning-2512-4bit | 4-bit | — | — | 223 |
| Ministral-3-8B-Instruct-2512-8bit | 8-bit | — | — | 170 |
| Ministral-3-8B-Instruct-2512-6bit | 6-bit | — | — | 128 |
| Ministral-3-8B-Reasoning-2512-8bit | 8-bit | — | — | 71 |
| Ministral-3-8B-Instruct-2512-bf16 | bf16 | — | — | 61 |
| Ministral-3-8B-Reasoning-2512-6bit | 6-bit | — | — | 56 |
| Ministral-3-8B-Instruct-2512-5bit | 5-bit | — | — | 48 |
| Ministral-3-8B-Reasoning-2512-5bit | 5-bit | — | — | 42 |