4-bit Open on Hugging Face
Parameters
8.3B
Weights (est.)
≈ 5.6 GB
Mac memory (est.)
16 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/Fara-7B-4bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- microsoft/Fara-7B (quantized)
- Vendor
- Microsoft · Phi
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- mit
- Downloads
- 35 in the last 30 days · 170 all time · 0 likes
- Published
- 2026-07-31 · updated 2026-07-31 (3mo ago)
- Languages
- en
- Tags
- qwen2_5_vl, multimodal, computer-use, web-agent, agent, vision-language
Other MLX builds of microsoft/Fara-7B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Fara-7B-4bit (this one) | 4-bit | ≈ 5.6 GB | 16 GB+ | 35 |
| Fara-7B-8bit | 8-bit | ≈ 9.4 GB | 16 GB+ | 40 |