search

facebookresearch / audiocraft

Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable music generation LM with textual and melodic conditioning.
MIT License
20.18k stars 2.01k forks source link

Model Crashes on HuggingFace because of Outdated NVIDIA driver! #331

Open rezamora opened 8 months ago

rezamora commented 8 months ago

We are running a Space on HuggingFace (Currently offline: https://huggingface.co/spaces/TulipAIs/Soundscapes) which is based on Meta's Audiocraft. This model is currently running on the A10G large GPU. The space was working fine for a long time but recently we keep getting the following RuntimeError:

RuntimeError: The NVIDIA driver on your system is too old (found version 11040). Please update your GPU driver by downloading and installing a new version from the URL: http://www.nvidia.com/Download/index.aspx Alternatively, go to: https://pytorch.org/ to install a PyTorch version that has been compiled with your version of the CUDA driver.

Can you please assist us in fixing this issue?

yoinked-h commented 8 months ago

you are probably on cuda 11.4 ? upgrade cuda to 12.1/11.8 and install pytorch anew

rezamora commented 8 months ago

you are probably on cuda 11.4 ? upgrade cuda to 12.1/11.8 and install pytorch anew

Thank you! It is on HuggingFace! I don't think we can install drivers or upgrade cuda on HF can we?!

yoinked-h commented 8 months ago

you're gonna have to use dockerfile magic