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 mini/d3_E640.mlpackage/Data/com.apple.CoreML/model.mlmodel from FluidInference/d3-coreml: direct link, hf CLI and curl.
- Browser
- Download file 3.65 MB
-
https://huggingface.co/FluidInference/d3-coreml/resolve/main/mini/d3_E640.mlpackage/Data/com.apple.CoreML/model.mlmodel
- Command line
-
hf download hf://FluidInference/d3-coreml/mini/d3_E640.mlpackage/Data/com.apple.CoreML/model.mlmodel
-
curl -L -o model.mlmodel https://huggingface.co/FluidInference/d3-coreml/resolve/main/mini/d3_E640.mlpackage/Data/com.apple.CoreML/model.mlmodel
3.65 MB
- Xet hash:
- 8bfda342e26e35446c248269fd27caa04eb02bbe889a6a3a13c3c5804b92f03f
- Size of remote file:
- 3.65 MB
- SHA256:
- 61c8e64e6d12fe92848e58bf1656fe2596fb4f08b60efa38c8402eb7ba0fc2c0
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.