UNetTemporalAttentionMultiply - ComfyUI Built-in Node Documentation

The UNetTemporalAttentionMultiply node applies multiplication factors to different types of attention mechanisms in a temporal UNet model. It modifies the model by adjusting the weights of self-attention and cross-attention layers, distinguishing between structural and temporal components. This allows fine-tuning of how much influence each attention type has on the model’s output.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	The input model to modify with attention multipliers	MODEL	Yes	-
`self_structural`	Multiplier for self-attention structural components (default: 1.0)	FLOAT	No	0.0 - 10.0
`self_temporal`	Multiplier for self-attention temporal components (default: 1.0)	FLOAT	No	0.0 - 10.0
`cross_structural`	Multiplier for cross-attention structural components (default: 1.0)	FLOAT	No	0.0 - 10.0
`cross_temporal`	Multiplier for cross-attention temporal components (default: 1.0)	FLOAT	No	0.0 - 10.0

Outputs

Output Name	Description	Data Type
`model`	The modified model with adjusted attention weights	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): b43c7c51cbc71349dd267c4abe1704899a394e205a60a13b45869f690d83a0f0

UNetSelfAttentionMultiply - ComfyUI Built-in Node Documentation

CLIPTextEncodeControlnet - ComfyUI Built-in Node Documentation

⌘I

​Inputs

​Outputs

Inputs

Outputs