Instructions to use rahulol/Hermit-MiniCPM5-2B-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use rahulol/Hermit-MiniCPM5-2B-MLX-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download rahulol/Hermit-MiniCPM5-2B-MLX-4bit --local-dir Hermit-MiniCPM5-2B-MLX-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Hermit MiniCPM5 2B (MLX 4-bit)
MLX 4-bit conversion of SonGT1/MiniCPM5-abliterated-2B (Apache-2.0) for the Hermit app on iPhone and Mac. Converted with mlx-lm (group size 64, 4.5 bits per weight). The tokenizer class is set to PreTrainedTokenizerFast for compatibility.
- Downloads last month
- -
Model size
3B params
Tensor type
U32
ยท
BF16 ยท
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support