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 nano/embeddings.f16 from FluidInference/d3-coreml: direct link, hf CLI and curl.
- Browser
- Download file 1.02 GB
-
https://huggingface.co/FluidInference/d3-coreml/resolve/main/nano/embeddings.f16
- Command line
-
hf download hf://FluidInference/d3-coreml/nano/embeddings.f16
-
curl -L -o embeddings.f16 https://huggingface.co/FluidInference/d3-coreml/resolve/main/nano/embeddings.f16
1.02 GB
- Xet hash:
- e65fe5046642cc1303c80eeb1dfda2e739821be0be9e3b8d5938ecb0c988f0f6
- Size of remote file:
- 1.02 GB
- SHA256:
- 8782b61738abba82d7a1c11acdad6614b975a4ed64f508edbb0d76cb5568b8b3
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.