Skip to main content
The SamplerCustom node is designed to provide a flexible and customizable sampling mechanism for various applications. It enables users to select and configure different sampling strategies tailored to their specific needs, enhancing the adaptability and efficiency of the sampling process.

Inputs

ParameterDescriptionData Type
modelThe ‘model’ input type specifies the model to be used for sampling, playing a crucial role in determining the sampling behavior and output.MODEL
add_noiseThe ‘add_noise’ input type allows users to specify whether noise should be added to the sampling process, influencing the diversity and characteristics of the generated samples.BOOLEAN
noise_seedThe ‘noise_seed’ input type provides a seed for the noise generation, ensuring reproducibility and consistency in the sampling process when adding noise.INT
cfgThe ‘cfg’ input type sets the configuration for the sampling process, allowing for fine-tuning of the sampling parameters and behavior.FLOAT
positiveThe ‘positive’ input type represents positive conditioning information, guiding the sampling process towards generating samples that align with specified positive attributes.CONDITIONING
negativeThe ‘negative’ input type represents negative conditioning information, steering the sampling process away from generating samples that exhibit specified negative attributes.CONDITIONING
samplerThe ‘sampler’ input type selects the specific sampling strategy to be employed, directly impacting the nature and quality of the generated samples.SAMPLER
sigmasThe ‘sigmas’ input type defines the noise levels to be used in the sampling process, affecting the exploration of the sample space and the diversity of the output.SIGMAS
latent_imageThe ‘latent_image’ input type provides an initial latent image for the sampling process, serving as a starting point for sample generation.LATENT

Outputs

ParameterDescriptionData Type
outputThe ‘output’ represents the primary result of the sampling process, containing the generated samples.LATENT
denoised_outputThe ‘denoised_output’ represents the samples after a denoising process has been applied, potentially enhancing the clarity and quality of the generated samples.LATENT
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub