How to use mlx-community/nanoLLaVA-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir nanoLLaVA-4bit mlx-community/nanoLLaVA-4bit
