VideoTriangleCFGGuidance - ComfyUI Built-in Node Documentation

The VideoTriangleCFGGuidance node applies a triangular classifier-free guidance (CFG) scaling pattern to a video model. It varies the conditioning scale over time using a triangular wave that oscillates between min_cfg and the model’s original conditioning scale. This creates a dynamic guidance pattern that can help improve video generation consistency and quality.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The video model to apply triangular CFG guidance to	MODEL	Yes	-
`min_cfg`	The minimum CFG scale value for the triangular pattern (default: 1.0). This parameter is shown in the advanced section of the node interface.	FLOAT	Yes	0.0 - 100.0 (step: 0.5, round: 0.01)

Outputs

Output Name	Description	Data Type
`model`	The modified model with triangular CFG guidance applied	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): 412d84d402f8c9a4852ee7b3f0ca0ab5650658fc26a37d10333a653e92e0294e

​Inputs

​Outputs

Inputs

Outputs