2-bit Open on Hugging Face
Parameters
15B
Weights (est.)
≈ 5.3 GB
Mac memory (est.)
8 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/Apriel-1.5-15b-Thinker-2bit-MLX --image photo.jpg --prompt "Describe this image"
Details
- Base model
- ServiceNow-AI/Apriel-1.5-15b-Thinker (quantized)
- Vendor
- ServiceNow-AI
- Task
- Vision + text
- Quantisation
- 2-bit
- Licence
- mit
- Downloads
- 39 in the last 30 days · 348 all time · 1 likes
- Published
- 2025-10-02 · updated 2025-10-03 (1y ago)
- Languages
- en
- Tags
- pixtral, apple-silicon, multimodal, vision-language, llava, quantized, 2bit
Other MLX builds of ServiceNow-AI/Apriel-1.5-15b-Thinker
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Apriel-1.5-15b-Thinker-2bit-MLX (this one) | 2-bit | ≈ 5.3 GB | 8 GB+ | 39 |
| Apriel-1.5-15b-Thinker-3bit-MLX | 3-bit | — | — | 43 |
| Apriel-1.5-15b-Thinker-6bit-MLX | 6-bit | — | — | 36 |