HunyuanVideo15SuperResolution - ComfyUI Built-in Node Documentation - ComfyUI

The HunyuanVideo15SuperResolution node prepares conditioning data for a video super-resolution process. It takes a latent representation of a video and, optionally, a starting image, and packages them along with noise augmentation and CLIP vision data into a format that can be used by a model to generate a higher-resolution output.

Inputs

Parameter	Description	Data Type	Required	Range
`positive`	The positive conditioning input to be modified with latent and augmentation data.	CONDITIONING	Yes	N/A
`negative`	The negative conditioning input to be modified with latent and augmentation data.	CONDITIONING	Yes	N/A
`vae`	The VAE used to encode the optional `start_image`. Required if `start_image` is provided.	VAE	No	N/A
`start_image`	An optional starting image to guide the super-resolution. If provided, it is upscaled and encoded into the conditioning latent.	IMAGE	No	N/A
`clip_vision_output`	Optional CLIP vision embeddings to add to the conditioning.	CLIP_VISION_OUTPUT	No	N/A
`latent`	The input latent video representation that is incorporated into the conditioning.	LATENT	Yes	N/A
`noise_augmentation`	The strength of noise augmentation to apply to the conditioning (default: 0.70). This is an advanced parameter.	FLOAT	No	0.0 - 1.0 (step 0.01)

Note: If you provide a start_image, you must also connect a vae so it can be encoded. The start_image is automatically upscaled to 16 times the spatial dimensions (width and height) of the input latent, then encoded and placed into the conditioning latent. Only the RGB channels of the start_image are used for encoding.

Outputs

Output Name	Description	Data Type
`positive`	The modified positive conditioning, now containing the concatenated latent, noise augmentation, and optional CLIP vision data.	CONDITIONING
`negative`	The modified negative conditioning, now containing the concatenated latent, noise augmentation, and optional CLIP vision data.	CONDITIONING
`latent`	The input latent is passed through 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): c9e64092e78423f5e0dc43446a77240e09100242c25e4fccc91491049fe76be5