Instructions to use moonbox-dev/MobileSAM-MLX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use moonbox-dev/MobileSAM-MLX with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download moonbox-dev/MobileSAM-MLX --local-dir MobileSAM-MLX
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download 8bit/config.json from moonbox-dev/MobileSAM-MLX: direct link, hf CLI and curl.
- Browser
- Download file 358 Bytes
-
https://huggingface.co/moonbox-dev/MobileSAM-MLX/resolve/main/8bit/config.json
- Command line
-
hf download hf://moonbox-dev/MobileSAM-MLX/8bit/config.json
-
curl -L -o config.json https://huggingface.co/moonbox-dev/MobileSAM-MLX/resolve/main/8bit/config.json
358 Bytes
| { | |
| "model_type": "mobile_sam", | |
| "image_size": 1024, | |
| "embed_dims": [ | |
| 64, | |
| 128, | |
| 160, | |
| 320 | |
| ], | |
| "depths": [ | |
| 2, | |
| 2, | |
| 6, | |
| 2 | |
| ], | |
| "num_heads": [ | |
| 2, | |
| 4, | |
| 5, | |
| 10 | |
| ], | |
| "window_sizes": [ | |
| 7, | |
| 7, | |
| 14, | |
| 7 | |
| ], | |
| "prompt_dim": 256, | |
| "mask_in_chans": 16, | |
| "quantization": { | |
| "group_size": 64, | |
| "bits": 8, | |
| "mode": "affine" | |
| }, | |
| "precision": "8bit" | |
| } | |