Instructions to use facebook/magnet-small-10secs with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Audiocraft
How to use facebook/magnet-small-10secs with Audiocraft:
from audiocraft.models import MAGNeT model = MAGNeT.get_pretrained("facebook/magnet-small-10secs") descriptions = ['disco beat', 'energetic EDM', 'funky groove'] wav = model.generate(descriptions) # generates 3 samples. - Notebooks
- Google Colab
- Kaggle
Download state_dict.bin from facebook/magnet-small-10secs: direct link, hf CLI and curl.
- Browser
- Download file 841 MB
-
https://huggingface.co/facebook/magnet-small-10secs/resolve/main/state_dict.bin
- Command line
-
hf download hf://facebook/magnet-small-10secs/state_dict.bin
-
curl -L -o state_dict.bin https://huggingface.co/facebook/magnet-small-10secs/resolve/main/state_dict.bin
841 MB
- Xet hash:
- 449986b415f9359cea7f3d2b7827e5d4a7204b2e50f0480a4d810cea6391726a
- Size of remote file:
- 841 MB
- SHA256:
- 0594e551ed9c40464b5918f5ddcce348e491e912e61d69f4d5d64d4ddd1a6ade
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.