
On this page
Short answer
On November 5, 2026, Google made Gemini Omni 1.2 Flash generally available. The video model can generate and edit clips up to 15 seconds long (up from 10 seconds in Omni 1.1), and you can refine the same clip in a multi-turn chat instead of starting over. In Google Vids, 15-second clips come at no additional cost with a Google account or Google Workspace. In the Gemini app, the feature is rolling out to Google AI Ultra subscribers. Outputs carry Google's SynthID watermark.
What happened
In a Google DeepMind product post, product manager Anish Nangia said Gemini Omni 1.2 Flash adds native 15-second video creation, conversational editing, and stronger reference following. You can set clip length in one-second steps from 3 to 15 seconds. Google also cut the API price by 30% from Omni 1.1, to $0.07 per second for 720p video, according to that post and a matching Google Cloud blog.
The workflow Google describes is draft fast at 360p or 720p, fix a detail with a follow-up edit, then upscale the keepers to 1080p or 4K. You can steer generation with character, product or style images, short motion clips up to three seconds, and (for allowlisted users) audio cues. Google says faces, garments, packaging type and brand colors hold more steadily across camera angles than before.
The Verge reports that Omni 1.2 Flash can make edits without regenerating the whole clip, and that Google claims better continuous camera moves and more natural-sounding dialogue. Google Cloud says the model is production-ready with Provisioned Throughput for business use, and that creative tools such as Artlist, Photoroom, Picsart and others are already wiring it in.
Where everyday users can try it, according to Google: direct scenes in Flow, make 15-second clips in Google Vids, create videos in the Gemini app (Ultra rollout), and use it for ads. Availability outside those products depends on the Gemini API and Google Cloud.
Why people are talking about it
Fifteen seconds is long enough for a short explainer, a product spin or a social spot in one generation — a concrete step up from the previous 10-second ceiling. Combined with chat-style edits and a free path in Google Vids, it is something many people who already use Gemini can test the same week, not only developers.
What it means if you use AI
- Video joins the everyday AI toolkit. If you already ask chatbots for drafts and images, short clips are now closer to the same loop: describe, tweak, keep.
- Editing in conversation cuts rework. Changing a background, a product or a camera angle on the same clip is closer to how people already talk to AI than a full re-render.
- It is still a draft. Check faces, logos, on-screen text and physics before you post anything public. Google notes that sports action, rapid script changes and complex physics still need work.
- Watermark and plan limits apply. Clips carry SynthID; Gemini-app access is tied to Google AI Ultra while Vids offers a free account path with Google's stated limits.
Quick checklist
- Open Google Vids with a Google account and look for 15-second AI clip creation.
- If you have Google AI Ultra, check whether 15-second video has rolled out in the Gemini app.
- Prefer a clear first prompt (length, subject, camera), then edit in follow-up turns.
- Review every frame for faces, logos and text before sharing.
- Developers: try the model in Google AI Studio or the Gemini API; note the 30% lower 720p price Google lists.
Sources
- Google DeepMind / The Keyword (Nov 5, 2026): Gemini Omni 1.2 Flash — Fast and cost-effective with longer clips (primary source)
- Google Cloud Blog (Nov 5, 2026): Gemini Omni 1.2 Flash now GA — cost-effective video
- The Verge (Nov 5, 2026): Google’s Gemini Omni 1.2 Flash can generate clips up to 15 seconds long
- Cover image: Woman sitting on a chair in a cafe using her smartphone to make a video call, by Shixart1985, Wikimedia Commons, CC BY 2.0
Written by our editorial team with the help of AI tools. Spotted a mistake? Tell us in the discussion.