Fix Stable Diffusion 3.x pooled prompt embedding with multiple images #12306
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Outdated suggestions cannot be applied.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.
What does this PR do?
When having more than one prompts and and also
num_images_per_prompt > 1, pooled prompt embeddings are repeated incorrectly. This leads to the pooled prompt embeddings getting assigned to the wrong images.While CLIP and T5 prompt embeddings are repeated along the sequence dimension, pooled CLIP embeddings are currently repeated along the batch dimension. This PR fixes this by repeating over the embedding dimension.
Corresponding fix for Flux.1: #9280
Before submitting
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@yiyixuxu @sayakpaul