Instructions to use nvidia/OpenMath-CodeLlama-13b-Python with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use nvidia/OpenMath-CodeLlama-13b-Python with NeMo:
# tag did not correspond to a valid NeMo domain.
- Notebooks
- Google Colab
- Kaggle
Download nemo_model/model_weights/model.decoder.layers.mlp.linear_fc2.weight/12.0.1 from nvidia/OpenMath-CodeLlama-13b-Python: direct link, hf CLI and curl.
- Browser
- Download file 35.4 MB
-
https://huggingface.co/nvidia/OpenMath-CodeLlama-13b-Python/resolve/main/nemo_model/model_weights/model.decoder.layers.mlp.linear_fc2.weight/12.0.1
- Command line
-
hf download hf://nvidia/OpenMath-CodeLlama-13b-Python/nemo_model/model_weights/model.decoder.layers.mlp.linear_fc2.weight/12.0.1
-
curl -L -o 12.0.1 https://huggingface.co/nvidia/OpenMath-CodeLlama-13b-Python/resolve/main/nemo_model/model_weights/model.decoder.layers.mlp.linear_fc2.weight/12.0.1
35.4 MB
- Xet hash:
- 5ea272ec2ec4b77db496d7b7427b0ecc50115149f735fa2ed2b8ad635f0dd51e
- Size of remote file:
- 35.4 MB
- SHA256:
- c805da73fe02ffc556a5be138c6bc1163853561c5e4832c00c00866cdf1c3951
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.