How to use nicolasramos/Qwen3.8-4B-Distill-MLX-oQ4e-fp16-mtp with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Qwen3.8-4B-Distill-MLX-oQ4e-fp16-mtp nicolasramos/Qwen3.8-4B-Distill-MLX-oQ4e-fp16-mtp
