Video · by Google

Gemini Omni

Change a video by chatting, like swapping out objects.

text → video · image, video, text → video · video, text → video
Way in
Official APISign up yourself, api key
To start
$$none (paid tier only)
Runs
In the cloudAlso through Vertex AI, fal, OpenRouter, runway, pika
Checked
Nov 3, 2026Prices as of that day

What it can do 4

Generate a video clip with native audio from a text prompt16:9 or 9:16; 360p/720p native, 1080p/4K upscaled; clips up to ~10 s per turntext → video~$0.10/s at 720p (Gemini API token billing: $17.50/M video-output tokens, ~5,792 tokens/s); fal: $0.03/s 360p, $0.10 720p, $0.15 1080p, $0.30 4K
Animate images / generate from first-last frame or subject/style reference images and up to 3 video referencesvideo refs: up to 3 clips of <=3 s each; audio in refs ignored; audio references not supportedimage, video, text → videosame per-second rate; input $1.50/M tokens
Edit a video conversationally (replace elements, change perspective) across turns via previous_interaction_iduploaded inputs <=10 s; editing uploads not in all regions; voice editing not supportedvideo, text → videoeach edit re-bills the full output clip
Extend a clip at its end, including continuing spoken dialogue3-10 s per turn, 40 s total; end-only; extending uploads blocked in EEA/CH/UKvideo, text → videoper output second

Does the same job

Prices and limits as of Nov 3, 2026, from Google’s own pages. They change; check before you rely on them.Not affiliated with Google. Not advice.Something wrong?
Made by Jasper Burns

A short note when I ship something new, about once a month. One list for all my projects.