4-bit Open on Hugging Face
Parameters
2.2B
Weights (est.)
≈ 1.5 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/SmolVLM-Instruct-4bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- HuggingFaceTB/SmolVLM-Instruct (finetune)
- Vendor
- Hugging Face
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 696 in the last 30 days · 14K all time · 6 likes
- Published
- 2024-11-26 · updated 2024-11-29 (1y ago)
- Languages
- en
- Tags
- idefics3
Other MLX builds of HuggingFaceTB/SmolVLM-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| SmolVLM-Instruct-4bit (this one) | 4-bit | ≈ 1.5 GB | 8 GB+ | 696 |
| SmolVLM-Instruct-6bit | 6-bit | ≈ 2.0 GB | 8 GB+ | 48 |
| SmolVLM-Instruct-8bit | 8-bit | ≈ 2.5 GB | 8 GB+ | 63 |
| SmolVLM-Instruct-bf16 | bf16 | ≈ 4.5 GB | 8 GB+ | 74 |