Fix black image output in QwenImageEdit docstring#13820
Open
Praanjaljain29 wants to merge 1 commit into
Open
Conversation
Fixes huggingface#13819 by upcasting the VAE to torch.float32 before decoding to prevent half-precision overflow.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #13819
What does this PR do?
This fixes a bug in the
QwenImageEditPipelinedocumentation snippet where users were getting a black image output.Because the pipeline was initialized in
torch.bfloat16, theAutoencoderKL(VAE) mathematically overflows during thedecodestep, resulting inNaNvalues that the image processor converts to black pixels.This PR adds
pipe.vae.to(torch.float32)to the docstring example to prevent the overflow and ensure proper image generation.Before submitting
Who can review?
@yiyixuxu @patrickvonplaten