Skip to main content
The DualCFGGuider node creates a guidance system for dual classifier-free guidance sampling. It combines two positive conditioning inputs with one negative conditioning input, applying different guidance scales to each conditioning pair to control the influence of each prompt on the generated output.

Inputs

ParameterDescriptionData TypeRequiredRange
modelThe model to use for guidanceMODELYes-
cond1The first positive conditioning inputCONDITIONINGYes-
cond2The second positive conditioning inputCONDITIONINGYes-
negativeThe negative conditioning inputCONDITIONINGYes-
cfg_condsGuidance scale for the first positive conditioning (default: 8.0)FLOATYes0.0 - 100.0
cfg_cond2_negativeGuidance scale for the second positive and negative conditioning (default: 8.0)FLOATYes0.0 - 100.0
styleThe guidance style to apply (default: “regular”). When set to “nested”, the guidance is applied in a nested fashionCOMBOYes”regular"
"nested”

Outputs

Output NameDescriptionData Type
GUIDERA configured guidance system ready for use with samplingGUIDER
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): 5440fb55429b8c30ef88ee0eff74c8a54916b1ce644f93d5d0d5742b5056637d