TemporalScoreRescaling - ComfyUI Built-in Node Documentation - ComfyUI

This node applies Temporal Score Rescaling (TSR) to a diffusion model. It modifies the model’s sampling behavior by rescaling the predicted noise or score during the denoising process, which can steer the diversity of the generated output. This is implemented as a post-CFG (Classifier-Free Guidance) function.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The diffusion model to be patched with the TSR function.	MODEL	Yes	-
`tsr_k`	Controls the rescaling strength. Lower k produces more detailed results; higher k produces smoother results in image generation. Setting k = 1 disables rescaling. (default: 0.95)	FLOAT	Yes	0.01 - 100.0
`tsr_sigma`	Controls how early rescaling takes effect. Larger values take effect earlier. (default: 1.0)	FLOAT	Yes	0.01 - 100.0

Outputs

Output Name	Description	Data Type
`patched_model`	The input model, now patched with the Temporal Score Rescaling function applied to its sampling process.	MODEL

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): 4d4e3c64fb6e3a3fe4725ea944a361b46d871943a10e65d72d70e0e6d757dfca