LTXVImgToVideoInplace - ComfyUI Built-in Node Documentation - ComfyUI

LTXVImgToVideoInplace encodes an input image into the latent space and places those encoded frames at the start of an existing latent video. The strength value controls how strongly the encoded image conditions those initial frames, and when bypass is enabled, the input latent is returned unchanged.

Inputs

Parameter	Description	Data Type	Required	Range
`vae`	The VAE model used to encode the input image into the latent space.	VAE	Yes	-
`image`	The input image to be encoded and used to condition the video latent.	IMAGE	Yes	-
`latent`	The target latent video representation to be modified.	LATENT	Yes	-
`strength`	Controls how strongly the encoded image conditions the initial frames of the latent. A value of 1.0 fully conditions the initial frames with the encoded image, while lower values condition them less strongly. The noise mask for the initial frames is set to `1.0 - strength`. (default: 1.0)	FLOAT	No	0.0 - 1.0
`bypass`	Bypass the conditioning. When enabled, the node returns the input latent unchanged. (default: False)	BOOLEAN	No	True or False

Note: The image will be automatically resized to match the spatial dimensions required by the vae for encoding, based on the latent input’s width and height. Only the RGB channels of the image are used for encoding.

Outputs

Output Name	Description	Data Type
`latent`	The resulting latent video representation. When bypass is disabled, it contains the updated `samples` and a `noise_mask` that applies the conditioning strength to the initial frames. When bypass is enabled, it is the input latent returned unchanged.	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): 69faa4b2e7b0fedeee531dc5a8809e23a79c9ce03e9760afb865160594fef30d