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 LaplaceScheduler node generates a sequence of sigma values following a Laplace distribution for use in diffusion sampling. It creates a schedule of noise levels that gradually decrease from a maximum to minimum value, using Laplace distribution parameters to control the progression. This scheduler is commonly used in custom sampling workflows to define the noise schedule for diffusion models.

Inputs

ParameterData TypeRequiredRangeDescription
stepsINTYes1 to 10000Number of sampling steps in the schedule (default: 20)
sigma_maxFLOATYes0.0 to 5000.0Maximum sigma value at the start of the schedule (default: 14.614642)
sigma_minFLOATYes0.0 to 5000.0Minimum sigma value at the end of the schedule (default: 0.0291675)
muFLOATYes-10.0 to 10.0Mean parameter for the Laplace distribution (default: 0.0)
betaFLOATYes0.0 to 10.0Scale parameter for the Laplace distribution (default: 0.5)

Outputs

Output NameData TypeDescription
SIGMASSIGMASA sequence of sigma values following a Laplace distribution schedule