Instructions to use nvidia/OpenMath-CodeLlama-7b-Python with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use nvidia/OpenMath-CodeLlama-7b-Python with NeMo:
# tag did not correspond to a valid NeMo domain.
- Notebooks
- Google Colab
- Kaggle
Download nemo_model/model_weights/model.decoder.final_layernorm.weight/0 from nvidia/OpenMath-CodeLlama-7b-Python: direct link, hf CLI and curl.
- Browser
- Download file 8.19 kB
-
https://huggingface.co/nvidia/OpenMath-CodeLlama-7b-Python/resolve/be1f8afe8a41feb106af3efa1f6be1e19c6fa5b3/nemo_model/model_weights/model.decoder.final_layernorm.weight/0
- Command line
-
hf download hf://nvidia/OpenMath-CodeLlama-7b-Python@be1f8afe8a41feb106af3efa1f6be1e19c6fa5b3/nemo_model/model_weights/model.decoder.final_layernorm.weight/0
-
curl -L -o 0 https://huggingface.co/nvidia/OpenMath-CodeLlama-7b-Python/resolve/be1f8afe8a41feb106af3efa1f6be1e19c6fa5b3/nemo_model/model_weights/model.decoder.final_layernorm.weight/0
8.19 kB
- Xet hash:
- fac517ebfc9a562940f55c5854355584c7cf72a4ba110604b07e8c71adef0394
- Size of remote file:
- 8.19 kB
- SHA256:
- a2ed4f64f0c536794046798ec6c85fc97b6d6f63a82824222e6a8fb0deb1c1b1
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.