Skip to main content
The FreeU node applies frequency-domain modifications to a model’s output blocks to enhance image generation quality. It works by scaling different channel groups and applying Fourier filtering to specific feature maps, allowing for fine-tuned control over the model’s behavior during the generation process.

Inputs

ParameterDescriptionData TypeRequiredRange
modelThe model to apply FreeU modifications toMODELYes-
b1Backbone scaling factor for model_channels × 4 features (default: 1.1)FLOATYes0.0 - 10.0
b2Backbone scaling factor for model_channels × 2 features (default: 1.2)FLOATYes0.0 - 10.0
s1Skip connection scaling factor for model_channels × 4 features (default: 0.9)FLOATYes0.0 - 10.0
s2Skip connection scaling factor for model_channels × 2 features (default: 0.2)FLOATYes0.0 - 10.0

Outputs

Output NameDescriptionData Type
modelThe modified model with FreeU patches appliedMODEL
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): a0f0b165638aa9e88ea415fc966bdac9bfbc13e9a5dcebc9780b31b63ce91f5c