Veo3VideoGenerationNode - ComfyUI Built-in Node Documentation - ComfyUI

Generates videos from text prompts using Google’s Veo 3 API. This node supports multiple Veo 3 models, including fast and lite variants, and allows you to specify video resolution, duration, and audio generation.

Inputs

Parameter	Description	Data Type	Required	Range
`prompt`	Text description of the video (default: "")	STRING	Yes	-
`aspect_ratio`	Aspect ratio of the output video (default: “16:9”)	COMBO	Yes	”16:9" "9:16”
`resolution`	Output video resolution. 4K is not available for the veo-3.1-lite model. (default: “720p”)	COMBO	No	”720p" "1080p" "4k”
`negative_prompt`	Negative text prompt to guide what to avoid in the video (default: "")	STRING	No	-
`duration_seconds`	Duration of the output video in seconds (default: 8)	INT	No	4 - 8 (step 2)
`enhance_prompt`	This parameter is deprecated and ignored. (default: True)	BOOLEAN	No	-
`person_generation`	Whether to allow generating people in the video (default: “ALLOW”)	COMBO	No	”ALLOW" "BLOCK”
`seed`	Seed for video generation (0 for random) (default: 0)	INT	No	0 - 4294967295
`image`	Optional reference image to guide video generation	IMAGE	No	-
`model`	Veo 3 model to use for video generation (default: “veo-3.1-generate”)	COMBO	No	”veo-3.1-generate" "veo-3.1-fast-generate" "veo-3.1-lite”
`generate_audio`	Generate audio for the video. Supported by all Veo 3 models. (default: False)	BOOLEAN	No	-

Note: The enhance_prompt parameter is deprecated and its value is ignored. The node always enhances the prompt internally. If you select “4k” resolution with the veo-3.1-lite model, the node raises an error.

Outputs

Output Name	Description	Data Type
`output`	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): 5320736448ad854e2f93e08ccaa870e977e06497666cb305f314bc76ff917740