Modelpublic

TyKaoz/gemma-4-12B-it-4bit

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face

0likes4downloads

Model Card

Gemma 4 12B Instruct — 4-bit MLX

4-bit MLX quantization of `google/gemma-4-12B-it`, for Apple Silicon (~6.3 GB). Vision-language model — run it with mlx-vlm, not mlx-lm.

bash

pip install -U mlx-vlm

bash

python -m mlx_vlm.generate \
  --model TyKaoz/gemma-4-12B-it-4bit \
  --prompt "Explique la quantization en une phrase." \
  --max-tokens 200

Base	Tool	Precision	Size
`google/gemma-4-12B-it`	`mlx-vlm`	4-bit · group 64	~6.3 GB

By [TyKaoz](https://www.tykaoz.bzh) — privacy-first native macOS LLM chat client. Apache 2.0, inherited from the base model.