WanDancerVideo - ComfyUI Built-in Node Documentation

The WanDancerVideo node prepares conditioning data and an empty latent tensor for video generation with the WanDancer model. It attaches optional starting images, masks, CLIP vision embeddings, and audio features to the positive and negative conditioning so they can guide the generated video.

Inputs

Parameter	Description	Data Type	Required	Range
`positive`	The positive conditioning to guide video generation.	CONDITIONING	Yes
`negative`	The negative conditioning to guide video generation.	CONDITIONING	Yes
`vae`	The VAE used to encode the start image into the latent space.	VAE	Yes
`width`	The width of the generated video in pixels (default: 480).	INT	Yes	16 to MAX_RESOLUTION (step: 16)
`height`	The height of the generated video in pixels (default: 832).	INT	Yes	16 to MAX_RESOLUTION (step: 16)
`length`	The number of frames in the generated video. Should stay 149 for WanDancer (default: 149).	INT	Yes	1 to MAX_RESOLUTION (step: 4)
`clip_vision_output`	The CLIP vision embeds for the first frame.	CLIP_VISION_OUTPUT	No
`clip_vision_output_ref`	The CLIP vision embeds for the reference image.	CLIP_VISION_OUTPUT	No
`start_image`	The initial image(s) to be encoded, can be any number of frames.	IMAGE	No
`mask`	Image conditioning mask for the start image(s). White is kept, black is generated. Used for the local generations.	MASK	No
`audio_encoder_output`	An audio encoder output that provides audio features, frame rate, and inject scale values, which are attached to the conditioning when provided.	AUDIO_ENCODER_OUTPUT	No

Notes on Parameter Behavior

start_image is optional. When provided, it is resized to width and height, encoded by the vae, and attached to both the positive and negative conditioning. If start_image has more frames than length, the extra frames are dropped. If it has fewer frames, the missing frames are filled with zero values.
mask only has an effect when start_image is also provided. White areas are kept, and black areas are generated.
clip_vision_output_ref only has an effect when clip_vision_output is also provided.
audio_encoder_output, when provided, attaches audio embeddings, frame rate, and injection scale to both the positive and negative conditioning.

Outputs

Output Name	Description	Data Type
`positive`	The positive conditioning with any attached start-image latent, mask, CLIP vision, or audio data.	CONDITIONING
`negative`	The negative conditioning with any attached start-image latent, mask, CLIP vision, or audio data.	CONDITIONING
`latent`	An empty latent tensor sized for the requested video 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): 086a0ec361cf7f7ae7ce9505b55d31d92b025c6c7c9cde192009e6664011ad05

​Inputs

​Notes on Parameter Behavior

​Outputs

Inputs

Notes on Parameter Behavior

Outputs