Translation
MLX
Safetensors
hy_v3
mlx-lm
apple-silicon
multilingual-translation
hy-v3
tencent-hy
quantized
4-bit precision
affine-quantization
Mixture of Experts
Instructions to use QwQbb/Hy-MT2-30B-A3B-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use QwQbb/Hy-MT2-30B-A3B-MLX-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Hy-MT2-30B-A3B-MLX-4bit QwQbb/Hy-MT2-30B-A3B-MLX-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
| { | |
| "backend": "tokenizers", | |
| "bos_token": "<|hy_begin▁of▁sentence|>", | |
| "clean_up_tokenization_spaces": false, | |
| "eos_token": "<eos:6124c78e>", | |
| "is_local": true, | |
| "local_files_only": false, | |
| "model_max_length": 1000000000000000019884624838656, | |
| "pad_token": "<|hy_▁pad▁|>", | |
| "tokenizer_class": "TokenizersBackend", | |
| "tool_parser_type": "glm47" | |
| } | |