Fix SD text encoder crash: infer sequence length from model#103
Merged
Conversation
StableDiffusionPipeline.encodeText() passed raw token count as the sequence dimension, but CLIP text encoders are exported with a fixed seq_len (77). When a prompt tokenizes to a different length, resolvingDynamicDimensions crashes. Fix: use CoreAITextEncoder.encode() which pads/truncates correctly, and infer the sequence length from the model input descriptor at load time instead of hardcoding 77. Fixes apple#102.
stikves
requested review from
alejandro-isaza,
carinapeng,
kevchengcodes and
tjia1818
alejandro-isaza
approved these changes
Jul 15, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 #102.
StableDiffusionPipeline.encodeText()passed raw token count as the sequence dimension, but CLIP text encoders are exported with a fixed seq_len (77 for SD 1.5/2.x). When a prompt tokenizes to a different length,resolvingDynamicDimensionscrashes with:Fix:
CoreAITextEncoder.encode()which already pads/truncates correctlyinferSequenceLength()helper onCoreAIDiffusionModelFunction