Apply x_embedder to controlnet to match hidden_states shape #691
+72
−2
Conversation
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
1 participant
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Outdated suggestions cannot be applied.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.
Motivation
Resolves #608, potentially other Controlnet related issues. After ComfyUI updated their Controlnet nodes it seems to have broken the Nunchaku implementation.
Nunchaku applies x_embedder to hidden_states in one of the forward calls but never applies the same to controlnet: https://github.com/nunchaku-tech/nunchaku/blob/ab324e25cc32420f36dfcab4d0f2180417f2d1d4/nunchaku/models/transformers/transformer_flux.py#L947
This mismatch doesnt get caught until the C++ FluxModel https://github.com/nunchaku-tech/nunchaku/blob/ab324e25cc32420f36dfcab4d0f2180417f2d1d4/src/FluxModel.cpp#L1286
Modifications
Ensure the controlnet tensor shapes match the expected hidden_states shape, and apply the x_embedder if applicable. Also adds some further assertions to fail earlier.
I imagine this would probably be better fixed at nunchaku/models/transformers/transformer_flux.py but I don't know diffusers well enough to test it outside of ComfyUI.