How to use mouddane/granite-speech-4.1-2b-nar-mlx-5bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir granite-speech-4.1-2b-nar-mlx-5bit mouddane/granite-speech-4.1-2b-nar-mlx-5bit
mouddane commited on 