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/embeddings.f16 from FluidInference/d3-coreml: direct link, hf CLI and curl.
- Browser
- Download file 1.27 GB
-
https://huggingface.co/FluidInference/d3-coreml/resolve/main/mini/embeddings.f16
- Command line
-
hf download hf://FluidInference/d3-coreml/mini/embeddings.f16
-
curl -L -o embeddings.f16 https://huggingface.co/FluidInference/d3-coreml/resolve/main/mini/embeddings.f16
1.27 GB
- Xet hash:
- b2a66968aa6b3bb200214e017f1392ad9b2375e9e6f2403a3e5e7c5b458ac11f
- Size of remote file:
- 1.27 GB
- SHA256:
- e41907b18d12ac35458d4df8606a51d59a58c0de0220383a6c3e5703ab4e0b08
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.