8-bit Open on Hugging Face
Parameters
4.6B
Weights (est.)
≈ 4.8 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/medgemma-4b-it-8bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- google/medgemma-4b-pt (finetune)
- Vendor
- Google · Gemma
- Task
- Vision + text
- Quantisation
- 8-bit
- Licence
- health-ai-developer-foundations
- Downloads
- 87 in the last 30 days · 1.5K all time · 3 likes
- Published
- 2025-05-21 · updated 2025-06-09 (1y ago)
- Tags
- gemma3, medical, radiology, clinical-reasoning, dermatology, pathology, ophthalmology, chest-x-ray
Other MLX builds of google/medgemma-4b-pt
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| medgemma-4b-it-4bit | 4-bit | ≈ 3.0 GB | 8 GB+ | 145 |
| medgemma-4b-it-6bit | 6-bit | ≈ 4.2 GB | 8 GB+ | 40 |
| medgemma-4b-it-8bit (this one) | 8-bit | ≈ 4.8 GB | 8 GB+ | 87 |
| medgemma-4b-it-bf16 | bf16 | ≈ 9.9 GB | 16 GB+ | 48 |