Instructions to use MERaLiON/MERaLiON-2-3B-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use MERaLiON/MERaLiON-2-3B-MLX with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download MERaLiON/MERaLiON-2-3B-MLX --local-dir MERaLiON-2-3B-MLX
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download adaptor_config.json from MERaLiON/MERaLiON-2-3B-MLX: direct link, hf CLI and curl.
- Browser
- Download file 82 Bytes
-
https://huggingface.co/MERaLiON/MERaLiON-2-3B-MLX/resolve/main/adaptor_config.json
- Command line
-
hf download hf://MERaLiON/MERaLiON-2-3B-MLX/adaptor_config.json
-
curl -L -o adaptor_config.json https://huggingface.co/MERaLiON/MERaLiON-2-3B-MLX/resolve/main/adaptor_config.json
82 Bytes
| { | |
| "speech_hidden_size": 1280, | |
| "text_hidden_size": 2304, | |
| "scale_factor": 15 | |
| } |