Skip to main content
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
TCFG (Tangential Damping CFG) implements a guidance technique that refines the unconditional (negative) predictions to better align with the conditional (positive) predictions. This method improves output quality by applying tangential damping to the unconditional guidance, based on the research paper referenced as 2503.18137. The node modifies the model’s sampling behavior by adjusting how unconditional predictions are processed during the classifier-free guidance process.

Inputs

ParameterData TypeRequiredRangeDescription
modelMODELYes-The model to apply tangential damping CFG to

Outputs

Output NameData TypeDescription
patched_modelMODELThe modified model with tangential damping CFG applied