google/gemini-omni-1.2-flash

Google's video generation and editing model with native audio. Makes 3 to 15 second clips from text, images, reference images, or an existing video.

Capabilities

Reference Images

Community model (estimated from hardware time)

Name	Type	Description	Default	Constraints
`prompt`*	string	Text prompt for video generation	`—`	—
`aspect_ratio`	string	Video aspect ratio. Ignored when editing a video	`"16:9"`	16:99:16
`duration`	integer	Video duration in seconds. Ignored when editing a video	`10`	min: 3, max: 15
`image`	string(uri)	Starting image to animate	`—`	—
`last_frame`	string(uri)	Ending image for interpolation. Requires image, up to 10 seconds	`—`	—
`reference_images`	array	Up to 6 images of characters, products, or styles to keep consistent		—
`resolution`	string	Resolution of the generated video. 1080p and 4k are upscaled	`"720p"`	360p720p1080p4k
`video`	string(uri)	Video to edit, up to 15 seconds. Describe the change in the prompt	`—`	—

promptrequiredstring

Text prompt for video generation

aspect_ratiostring

Video aspect ratio. Ignored when editing a video

Default: "16:9"

16:99:16

durationinteger

Video duration in seconds. Ignored when editing a video

Default: 10min: 3, max: 15

imagestring

Starting image to animate

last_framestring

Ending image for interpolation. Requires image, up to 10 seconds

reference_imagesarray

Up to 6 images of characters, products, or styles to keep consistent

Default:

resolutionstring

Resolution of the generated video. 1080p and 4k are upscaled

Default: "720p"

360p720p1080p4k

videostring

Video to edit, up to 15 seconds. Describe the change in the prompt

Version: 34bf4fcea846Updated: 11/7/2026300 runs