4-bit Open on Hugging Face
Parameters
24B
Weights (est.)
≈ 13.3 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-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/Devstral-Small-2505-4bit --prompt "Hello"
Details
- Base model
- mistralai/Devstral-Small-2505 (quantized)
- Vendor
- Mistral
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 304 in the last 30 days · 173K all time · 3 likes
- Published
- 2025-05-21 · updated 2025-05-21 (1y ago)
- Languages
- en, fr, de, es, pt, it, ja, ko, ru, zh, ar, fa, id, ms, ne, pl, ro, sr, sv, tr, uk, vi, hi, bn
- Tags
- mistral
Other MLX builds of mistralai/Devstral-Small-2505
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Devstral-Small-2505-3bit | 3-bit | ≈ 10.3 GB | 16 GB+ | 277 |
| Devstral-Small-2505-4bit (this one) | 4-bit | ≈ 13.3 GB | 24 GB+ | 304 |
| Devstral-Small-2505-6bit | 6-bit | ≈ 19.2 GB | 24 GB+ | 230 |
| Devstral-Small-2505-8bit | 8-bit | ≈ 25.0 GB | 32 GB+ | 224 |
| Devstral-Small-2505-bf16 | bf16 | ≈ 47.1 GB | 64 GB+ | 180 |