8-bit Open on Hugging Face
Parameters
—
Weights (est.)
—
Mac memory (est.)
—
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/gemma-3n-E2B-it-8bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- google/gemma-3n-E4B-it (quantized)
- Vendor
- Google · Gemma
- Task
- Vision + text
- Quantisation
- 8-bit
- Licence
- gemma
- Downloads
- 40 in the last 30 days · 417 all time · 0 likes
- Published
- 2025-07-06 · updated 2025-07-12 (1y ago)
- Tags
- gemma3n, automatic-speech-recognition, automatic-speech-translation, audio-text-to-text, video-text-to-text
Other MLX builds of google/gemma-3n-E4B-it
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| gemma-3n-E4B-it-lm-4bit | 4-bit | ≈ 3.9 GB | 8 GB+ | 1.5K |
| gemma-3n-E2B-it-3bit | 3-bit | ≈ 3.9 GB | 8 GB+ | 54 |
| gemma-3n-E2B-it-4bit | 4-bit | ≈ 4.5 GB | 8 GB+ | 1.2K |
| gemma-3n-E2B-it-5bit | 5-bit | ≈ 5.0 GB | 8 GB+ | 56 |
| gemma-3n-E4B-it-text-4bit-dwq | 4-bit DWQ | ≈ 9.6 GB | 16 GB+ | 46 |
| gemma-3n-E2B-it-bf16 | bf16 | ≈ 10.9 GB | 16 GB+ | 76 |
| gemma-3n-E4B-it-lm-bf16 | bf16 | ≈ 13.7 GB | 24 GB+ | 91 |
| gemma-3n-E2B-it-8bit (this one) | 8-bit | — | — | 40 |
| gemma-3n-E2B-it-6bit | 6-bit | — | — | 24 |