Wan22FunControlToVideo - ComfyUI Built-in Node Documentation

The Wan22FunControlToVideo node prepares conditioning data and an empty latent tensor for video generation with the Wan video model. It encodes optional reference images and control videos into latent space, attaches them to the positive and negative conditioning, and creates a zero-filled latent tensor with the correct spatial and temporal dimensions for the requested video.

Inputs

Parameter	Description	Data Type	Required	Range
`positive`	Positive conditioning input for guiding the video generation	CONDITIONING	Yes	-
`negative`	Negative conditioning input for guiding the video generation	CONDITIONING	Yes	-
`vae`	VAE model used to encode images into latent space	VAE	Yes	-
`width`	Width of the output video in pixels (default: 832, step: 16)	INT	Yes	16 to MAX_RESOLUTION
`height`	Height of the output video in pixels (default: 480, step: 16)	INT	Yes	16 to MAX_RESOLUTION
`length`	Number of frames in the video sequence (default: 81, step: 4)	INT	Yes	1 to MAX_RESOLUTION
`batch_size`	Number of video sequences to generate (default: 1)	INT	Yes	1 to 4096
`ref_image`	Optional reference image that provides visual guidance for the generation	IMAGE	No	-
`control_video`	Optional control video that guides the generation process	IMAGE	No	-

Note: The length parameter is processed in steps of 4 frames, and the node automatically applies temporal scaling when building the latent space. When ref_image is provided, only its first frame is encoded and attached to the conditioning as reference latents. When control_video is provided, it is trimmed to length frames, encoded, and placed into the concat latent used by the conditioning. The start_image parameter is referenced in the execution logic but is not exposed in the node’s input schema.

Outputs

Output Name	Description	Data Type
`positive`	Positive conditioning with video-specific latent data added, including the concat latent, mask, and optional reference latents	CONDITIONING
`negative`	Negative conditioning with video-specific latent data added, including the concat latent, mask, and optional reference latents	CONDITIONING
`latent`	Empty latent tensor prepared for video generation, sized according to the batch size, latent channels, length, height, and width	LATENT

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): 731b848f15c13ddc662f19230acb55d195f934bad7d9ae516a288e0ed8f8d899

​Inputs

​Outputs

Inputs

Outputs