WanSCAILToVideo - ComfyUI Built-in Node Documentation

The WanSCAILToVideo node prepares conditioning and an empty latent space for video generation with SCAIL and SCAIL-2 video models.

The WanSCAILToVideo node prepares conditioning and an empty latent space for video generation with SCAIL and SCAIL-2 video models. It processes optional inputs like reference images, pose videos, CLIP vision outputs, colored identity masks, and previous frame chunks, embedding them into the positive and negative conditioning. The node outputs the modified conditioning and a blank latent tensor of the specified video dimensions, ready for sampling.

Inputs

Parameter	Description	Data Type	Required	Range
positive	The positive conditioning input.	CONDITIONING	Yes	-
negative	The negative conditioning input.	CONDITIONING	Yes	-
vae	The VAE model used for encoding images and video frames.	VAE	Yes	-
width	The width of the output video in pixels (default: 512). Values step by 32.	INT	Yes	32 to MAX_RESOLUTION
height	The height of the output video in pixels (default: 896). Values step by 32.	INT	Yes	32 to MAX_RESOLUTION
length	The number of frames in the video (default: 81). Values step by 4.	INT	Yes	1 to MAX_RESOLUTION
batch_size	The number of videos to generate in a batch (default: 1).	INT	Yes	1 to 4096
pose_video	Video used for pose conditioning. Will be downscaled to half the resolution of the main video.	IMAGE	No	-
pose_video_mask	SCAIL-2 only. Colored per-identity SAM3 mask video at the same resolution as pose_video.	IMAGE	No	-
replacement_mode	SCAIL-2 only. False = Animation Mode (pose_video_mask should have black background). True = Replacement Mode (pose_video_mask should have white background). (default: False)	BOOLEAN	No	-
pose_strength	Strength of the pose latent. (default: 1.0)	FLOAT	Yes	0.0 to 10.0
pose_start	Start step of the pose conditioning. (default: 0.0)	FLOAT	Yes	0.0 to 1.0
pose_end	End step of the pose conditioning. (default: 1.0)	FLOAT	Yes	0.0 to 1.0
reference_image	Reference image. The first image is the primary reference (composite all identities onto it). SCAIL-2: extra batch images are used as additional views (back view, close-up, occluded background), each needing a matching reference_image_mask in that identity’s color.	IMAGE	No	-
reference_image_mask	SCAIL-2 only. Colored reference mask, batch matching reference_image (first = primary reference mask, rest = identity masks for the additional reference_image).	IMAGE	No	-
clip_vision_output	CLIP vision features for conditioning. Model is trained with stretch resize to aspect ratio.	CLIP_VISION_OUTPUT	No	-
video_frame_offset	Cumulative output frame this chunk begins at. Wire from the previous chunk’s video_frame_offset output. (default: 0)	INT	Yes	0 to MAX_RESOLUTION
previous_frame_count	Tail frames of previous_frames to anchor. SCAIL-2 trained at 5 (81-frame chunks, 76-frame step). (default: 5)	INT	Yes	1 to MAX_RESOLUTION
previous_frames	SCAIL-2 only. Full decoded output of the previous chunk. Only the last previous_frame_count are used as the extension anchor.	IMAGE	No	-

Note: The pose_video and pose_video_mask inputs are truncated together to the shorter of the two, and are processed only for the first length frames. If either input is shorter than or equal to video_frame_offset, it is ignored entirely. The pose_video is downscaled to half the resolution of the main video before encoding, and the encoded pose latent is multiplied by pose_strength and applied to the conditioning only between the pose_start and pose_end timestep steps. If pose_video_mask is provided, the colored mask video is downscaled to half resolution and converted into a 28-channel driving mask, which is added to both positive and negative conditioning.

Note: When reference_image is provided, each image in the batch is encoded individually into a latent and embedded into both positive and negative conditioning. The first image is the primary reference; additional images are used as additional views, each needing a matching reference_image_mask. reference_image_mask is only used when reference_image is also provided; when both are given, a 28-channel reference mask that binds the reference frames to identities is also built from the masks and added to the conditioning. In Replacement Mode (replacement_mode=True), the reference image is composited on a black background using the reference image mask as an alpha matte. When clip_vision_output is provided, it is applied to both positive and negative conditioning.

Note: When previous_frames is provided, only the last previous_frame_count frames are used as the extension anchor, and video_frame_offset is adjusted accordingly (reduced by the number of anchored frames, clamped at 0). The anchored frames are encoded and written into the beginning of the output latent, and a noise mask is included so those frames are kept unchanged during generation.

Outputs

Output Name	Description	Data Type
positive	The modified positive conditioning, potentially containing embedded reference image latents, CLIP vision output, pose video latents, driving masks, reference masks, or previous frame latents.	CONDITIONING
negative	The modified negative conditioning, potentially containing embedded reference image latents, CLIP vision output, pose video latents, driving masks, reference masks, or previous frame latents.	CONDITIONING
latent	An empty latent tensor of shape [batch_size, 16, ((length - 1) // 4) + 1, height // 8, width // 8]. When previous_frames is provided, the latent is partially filled with encoded previous frames and a noise mask is included.	LATENT
video_frame_offset	Adjusted offset + length. Wire into the next chunk for sequential video generation.	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): 4a1a2201dfa94bd2f1330db02ec18a5e0a6aae9e9ac5ae97d456b7af1aa84b7b