How to use EigenLabs/Qwen3.8-27B-MTP-bf16 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Qwen3.8-27B-MTP-bf16 EigenLabs/Qwen3.8-27B-MTP-bf16
