Vidu4ReferenceVideoNode - ComfyUI Built-in Node Documentation - ComfyUI

Generate a video from reference images, optional reference audio, and a prompt with a Vidu Q4 model. This is the reference-to-video variant of the Vidu Q4 generation nodes. Selecting a model reveals the parameters specific to that model.

Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use for video generation. Selecting a model reveals the parameters specific to it: `reference_images`, `reference_audios`, `prompt`, `aspect_ratio`, `resolution`, `duration`, `audio`, and `seed`.	DYNAMIC_COMBO	Yes	`"Vidu Q4 Preview"`
`reference_images`	Growable slot: connect one or more reference images (`image_1`, `image_2`, …) for the generated video; every image of a batch counts toward the total. Refer to them in the prompt by order: image 1, image 2, and so on.	IMAGE	Yes	Up to 15 images
`reference_audios`	Growable slot: connect optional voice references (`audio_1`, `audio_2`, `audio_3`), 3 to 12 seconds each. Only the voice is used, not the words: write the dialogue in the prompt and assign a voice by order, for example `image 1 says "Hello!" in the voice from audio 1`. Requires `audio` to be enabled.	AUDIO	No	Up to 3 clips
`prompt`	A textual description for video generation, up to 5000 characters. Needed to describe the references you want to use.	STRING	Yes	Any text
`aspect_ratio`	The aspect ratio of the output video.	COMBO	Yes	`"16:9"` `"9:16"` `"1:1"` `"3:4"` `"4:3"`
`resolution`	Resolution of the output video (default: `"720p"`).	COMBO	Yes	`"540p"` `"720p"` `"1080p"` `"2K"` `"4K"`
`duration`	Duration of the output video in seconds (default: 5).	INT	Yes	3 to 16
`audio`	When enabled, outputs video with sound, including dialogue and sound effects (default: True).	BOOLEAN	Yes	`True` `False`
`seed`	Seed controls whether the node should re-run; results are non-deterministic regardless of seed. This parameter has “control after generate” functionality (default: 42).	INT	Yes	1 to 2147483647

Note: At most 15 reference images can be used in total, counting every image of a batch. Each image must be at least 128x128 pixels with an aspect ratio between 1:5 and 5:1. Reference audio requires audio to be enabled and raising an error otherwise.

Outputs

Output Name	Description	Data Type
`VIDEO`	The generated video file.	VIDEO

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): f37ceec93a6140d69332415b8fd748d55e6a608177430975b4b42ab33e593489