Skip to main content
The SamplerLCMUpscale node provides a specialized sampling method that combines Latent Consistency Model (LCM) sampling with image upscaling capabilities. It allows you to upscale images during the sampling process using various interpolation methods, making it useful for generating higher resolution outputs while maintaining image quality.

Inputs

ParameterDescriptionData TypeRequiredRange
scale_ratioThe scaling factor to apply during upscaling (default: 1.0)FLOATNo0.1 - 20.0
scale_stepsThe number of steps to use for upscaling process. Use -1 for automatic calculation (default: -1)INTNo-1 - 1000
upscale_methodThe interpolation method used for upscaling the imageCOMBOYes”bislerp"
"nearest-exact"
"bilinear"
"area"
"bicubic”

Outputs

Output NameDescriptionData Type
samplerReturns a configured sampler object that can be used in the sampling pipelineSAMPLER
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): 11cd4de455d8345f0008757e7402b6cba1c21cf6eeebdd41f4500e615d826920