Add text_encoder_dtype and compile_text_encoder config parameters for Wan text encoders.#408
Open
ninatu wants to merge 1 commit into
Open
Add text_encoder_dtype and compile_text_encoder config parameters for Wan text encoders.#408ninatu wants to merge 1 commit into
text_encoder_dtype and compile_text_encoder config parameters for Wan text encoders.#408ninatu wants to merge 1 commit into
Conversation
… for Wan text encoders. Mitigates PR#397 by enabling the text encoder dtype to be configured separately from `weights_dtype`. Separating these parameters makes it possible to compile/load the text encoder in float32 (as it was before #397 by default) while keeping the model weights in bfloat16, and provides a configuration parameter to make the text encoder compilation optional (as it wasn’t compiled before #397). This addresses problems since some environments experience issues when `torch.compile` is run on it in hermetic/packaged setups.
ninatu
changed the title
text_encoder_dtype and compile_text_encoder config parameters…text_encoder_dtype and compile_text_encoder config parameters for Wan text encoders.
Perseus14
approved these changes
May 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Mitigates #397 by enabling the text encoder dtype to be configured separately from
weights_dtype. Separating these parameters makes it possible to compile/load the text encoder in float32 (as it was before #397 by default) while keeping the model weights in bfloat16, and provides a configuration parameter to make the text encoder compilation optional (as it wasn’t compiled before #397).This addresses problems since some environments experience issues when
torch.compileis run on it in hermetic/packaged setups.