How to use wezzel98765/gemma-4-12B-it-oQ8-fp16 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir gemma-4-12B-it-oQ8-fp16 wezzel98765/gemma-4-12B-it-oQ8-fp16
murphy1296