This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHubThe 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
| Parameter | Data Type | Required | Range | Description |
|---|---|---|---|---|
model | MODEL | Yes | - | The model to apply FreeU modifications to |
b1 | FLOAT | Yes | 0.0 - 10.0 | Backbone scaling factor for model_channels × 4 features (default: 1.1) |
b2 | FLOAT | Yes | 0.0 - 10.0 | Backbone scaling factor for model_channels × 2 features (default: 1.2) |
s1 | FLOAT | Yes | 0.0 - 10.0 | Skip connection scaling factor for model_channels × 4 features (default: 0.9) |
s2 | FLOAT | Yes | 0.0 - 10.0 | Skip connection scaling factor for model_channels × 2 features (default: 0.2) |
Outputs
| Output Name | Data Type | Description |
|---|---|---|
model | MODEL | The modified model with FreeU patches applied |