Skip to content

Conversation

@YFrisch
Copy link

@YFrisch YFrisch commented Jan 21, 2026

Update checkpoint saving to use raw_model instead of model.

Allows the use of multi-GPU training of the VAE with PyTorch DataParallel.

Update checkpoint saving to use raw_model.

Allows the use of multi-GPU training of the VAE with PyTorch DataParallel.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant