Instructions to use FluidInference/d3-coreml with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use FluidInference/d3-coreml with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download FluidInference/d3-coreml --local-dir d3-coreml
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download edge/vision_P3072.mlpackage/Data/com.apple.CoreML/model.mlmodel from FluidInference/d3-coreml: direct link, hf CLI and curl.
- Browser
- Download file 153 kB
-
https://huggingface.co/FluidInference/d3-coreml/resolve/main/edge/vision_P3072.mlpackage/Data/com.apple.CoreML/model.mlmodel
- Command line
-
hf download hf://FluidInference/d3-coreml/edge/vision_P3072.mlpackage/Data/com.apple.CoreML/model.mlmodel
-
curl -L -o model.mlmodel https://huggingface.co/FluidInference/d3-coreml/resolve/main/edge/vision_P3072.mlpackage/Data/com.apple.CoreML/model.mlmodel
153 kB
- Xet hash:
- 20f2ca9ecfb7d4c9487dd8306c1bd630f33d751f6e8e1b58ff1adc50b30a6a8a
- Size of remote file:
- 153 kB
- SHA256:
- 8fcae2dbf0259bce8440f2c1ca1d6e3589621a422b3c2f46b6f1a66d31ced3d4
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.