MiniMaxH3AddGuide - ComfyUI Built-in Node Documentation

This node anchors an image, a short clip, audio, or a clip with its soundtrack at any chosen frame of a MiniMax H3 video. It adds a guide keyframe to the conditioning at the specified frame index, and you can chain several of these nodes to anchor several frames in the same video.

Inputs

Parameter	Description	Data Type	Required	Range
`positive`	The conditioning to which the guide keyframe is attached.	CONDITIONING	Yes	-
`latent`	The MiniMax H3 audio-video latent that defines the target video. Must be a MiniMax H3 AV latent (nested, with two 5D tensors of 24 channels each).	LATENT	Yes	-
`frame_idx`	Frame index to anchor the image or the clip’s first frame at. Negative values are counted from the end of the video. (default: 0)	INT	Yes	-9999 to 9999
`vae`	Video VAE, needed when an image is connected.	VAE	No	-
`audio_vae`	Audio VAE, needed when an audio is connected.	VAE	No	-
`image`	Image or video frames to anchor. Multi-frame batches are anchored as a clip and cropped down to the model’s valid clip lengths: 5, 22, 39… (17k + 5) frames. Batches shorter than 5 frames use only the first image.	IMAGE	No	-
`audio`	Soundtrack to anchor starting at the same frame index, cropped to the video’s remaining duration.	AUDIO	No	-

Constraints:

At least one of image or audio must be provided; otherwise the node raises an error.
vae is required when image is connected.
audio_vae is required when audio is connected.
image batches with fewer than 5 frames use only the first image; batches of 5 or more frames are cropped down to a valid clip length (5, 22, 39, etc.).
frame_idx must place the guide within the video’s frame range, and a multi-frame clip must fit entirely in the video; otherwise the node raises an error.
When audio is connected, the frame index must not be past the end of the video’s audio track.

Outputs

Output Name	Description	Data Type
`positive`	The conditioning with the guide keyframe added, containing the resolved frame index and, when provided, the encoded image or audio latents.	CONDITIONING

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): 7a2f742421cc2655bd9c914258801e4538f1554a7c5e2b0836b2df1577f5a104

​Inputs

​Outputs

Inputs

Outputs