MLX
Safetensors
glm4_moe_lite_mtp
speculative-decoding
multi-token-prediction
drafter
4-bit precision
Instructions to use samithaj/GLM-4.7-Flash-MTP-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use samithaj/GLM-4.7-Flash-MTP-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir GLM-4.7-Flash-MTP-4bit samithaj/GLM-4.7-Flash-MTP-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
| { | |
| "_from_model_config": true, | |
| "eos_token_id": [ | |
| 154820, | |
| 154827, | |
| 154829 | |
| ], | |
| "pad_token_id": 154820, | |
| "temperature": 1.0, | |
| "transformers_version": "5.0.0.dev0" | |
| } | |