Skip to main content
The WanVaceToVideo node processes video conditioning data for video generation models. It takes positive and negative conditioning inputs along with video control data and prepares latent representations for video generation. The node handles video upscaling, masking, and VAE encoding to create the appropriate conditioning structure for video models.

Inputs

ParameterDescriptionData TypeRequiredRange
positivePositive conditioning input for guiding the generationCONDITIONINGYes-
negativeNegative conditioning input for guiding the generationCONDITIONINGYes-
vaeVAE model used for encoding images and video framesVAEYes-
widthOutput video width in pixels (default: 832, step: 16)INTYes16 to MAX_RESOLUTION
heightOutput video height in pixels (default: 480, step: 16)INTYes16 to MAX_RESOLUTION
lengthNumber of frames in the video (default: 81, step: 4)INTYes1 to MAX_RESOLUTION
batch_sizeNumber of videos to generate simultaneously (default: 1)INTYes1 to 4096
strengthControl strength for video conditioning (default: 1.0, step: 0.01)FLOATYes0.0 to 1000.0
control_videoOptional input video for control conditioning. If not provided, a neutral gray video is created automatically.IMAGENo-
control_masksOptional masks for controlling which parts of the video to modify. If not provided, a full white mask is used.MASKNo-
reference_imageOptional reference image for additional conditioning. When provided, it is encoded and prepended to the latent sequence.IMAGENo-
Note: When control_video is provided, it will be upscaled to match the specified width and height. If control_masks are provided, they must match the dimensions of the control video. The reference_image is encoded through the VAE and prepended to the latent sequence when provided. The length parameter determines the number of frames, and the latent length is calculated as ((length - 1) // 4) + 1.

Outputs

Output NameDescriptionData Type
positivePositive conditioning with video control data (vace_frames, vace_mask, vace_strength) appliedCONDITIONING
negativeNegative conditioning with video control data (vace_frames, vace_mask, vace_strength) appliedCONDITIONING
latentEmpty latent tensor ready for video generation with shape [batch_size, 16, latent_length, height/8, width/8]LATENT
trim_latentNumber of latent frames to trim when reference image is used (0 if no reference image is provided)INT
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): e31638510efa11e35f297becb4a9f070fdb84d34878868aaf3525e589e5abb0b