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
The SamplerEulerAncestralCFGPP node creates a specialized sampler for generating images using the Euler Ancestral method with classifier-free guidance. This sampler combines ancestral sampling techniques with guidance conditioning to produce diverse image variations while maintaining coherence. It allows fine-tuning of the sampling process through parameters that control noise and step size adjustments.

Inputs

ParameterData TypeRequiredRangeDescription
etaFLOATYes0.0 - 1.0Controls the step size during sampling, with higher values resulting in more aggressive updates (default: 1.0)
s_noiseFLOATYes0.0 - 10.0Adjusts the amount of noise added during the sampling process (default: 1.0)

Outputs

Output NameData TypeDescription
samplerSAMPLERReturns a configured sampler object that can be used in the image generation pipeline