Key takeaway
On November 5, 2026, Google made Gemini Omni 1.2 Flash generally available, extending native clip length from 10 to 15 seconds and cutting 720p pricing by 30% to $0.07 per second.
What happened
Google published Gemini Omni 1.2 Flash on its official The Keyword blog on November 5, 2026, and said the model is generally available on the Gemini API and Google Cloud. Google describes this as the first time the Gemini Omni line has reached general availability, backed by Google Cloud's enterprise infrastructure and security.
A companion post on the Google Workspace blog said Google Vids users can generate 15-second clips at no additional cost with a Google account or Google Workspace subscription. Beyond Vids, the model also ships in Gemini for business, Google Flow, the Gemini app and Google Ads; in the Gemini app the capability is rolling out to Google AI Ultra subscribers.
Key facts
| Item | Gemini Omni 1.2 Flash | Change from 1.1 |
|---|---|---|
| Clip length | Native up to 15 seconds; duration settable from 3 to 15 seconds in one-second steps | 1.1 topped out at 10 seconds |
| 720p price | $0.07 per second | 30% lower |
| Reference inputs | Character, product and style images, motion clips up to three seconds, audio cues | Google says reference adherence improved |
| Editing | Multi-turn conversational edits for product, background, lighting, weather and aspect ratio | Previously required regenerating the clip |
| Resolution workflow | Draft at 360p or 720p, then upscale selected cuts to 1080p or 4K | Now spelled out as the official workflow |
| Provenance | Outputs carry the imperceptible SynthID watermark | Consistent with the Omni line |
| Enterprise | Provisioned Throughput available on Google Cloud | Added with general availability |
At the published rate, a 15-second 720p generation costs $1.05 (0.07 x 15). That figure is calculated by this site from Google's stated unit price.
Capability details
Google lists four changes.
First, reference adherence. Google says faces, garments, packaging typography and brand colors stay consistent across camera angles. Given one model photo plus front-and-back product shots of three outfits, the model generates a 15-second virtual try-on sequence across 360-degree turns and match cuts while keeping the model's identity and each garment's cut, drape and back details.
Second, multi-turn editing. Users can refine the same 15-second clip turn by turn, swapping a product or background, changing lighting or weather, or switching aspect ratio, without re-briefing or regenerating. Google's example starts from one generated clip and swaps the background through a striped pavilion tent, an oak forest, a stable with a white horse, a market fair and a stone castle while keeping both characters' performances, armor reflections and camera framing.
Third, speech and camera control. Google says dialogue sounds more natural, lip-sync is clearer and multi-speaker conditioning is supported. On the camera side, the model can execute continuous moves such as FPV flythroughs, dramatic zoom-outs and orbits with fewer unwanted jump cuts.
Fourth, first-to-last frame consistency and scene extension. The model can interpolate between a start and end frame, or extend a clip into a longer sequence, with products, characters, props and environments staying consistent from shot to shot.
Google also states current limits: fast sports action, rapid script changes and complex physics are still being improved.
Industry impact
The verifiable facts: Adobe has integrated Gemini Omni Flash into Firefly and is adding version 1.2 for longer outputs and stronger reference consistency. Whering, Monks, Figma Weave, Magnific and STAN are listed by Google as early users; Whering uses it for virtual try-on and STAN turns still reference images into vertical 15-second festival promos and commerce ads.
In this site's reading of the public information, 15-second clips, multi-turn editing and $0.07 per second together push the cost of one deliverable short-form ad asset into the one-dollar range. That changes the iteration economics for small advertisers and e-commerce creative teams: the old model priced each asset as a fresh generation, while the new workflow prices edits on the same clip. This is our assessment based on Google's published price and workflow, not a Google statement.
A second observable shift is product scope. In its May 2026 I/O announcement, Google positioned Gemini Omni as a model that can generate any output from any input, starting with video. General availability of 1.2 Flash moves that line from preview to a stage where teams can procure it at scale.
How to access
Google lists four routes.
First, developers call the Gemini API, prototype in Google AI Studio, and use the interactive starter apps, developer documentation and pricing page.
Second, businesses use the Google Cloud console and can reserve Provisioned Throughput.
Third, anyone with a Google account or Google Workspace subscription can generate 15-second clips in Google Vids at no additional cost.
Fourth, 15-second video in the Gemini app is rolling out to Google AI Ultra subscribers; Google Ads can scale video campaigns, and Flow supports multi-turn scene editing.
All outputs carry the SynthID watermark, which Google describes as imperceptible and usable to verify that content was AI-generated.
Official sources
- Google The Keyword: Gemini Omni 1.2 Flash: 15-second video and lower pricing (November 5, 2026)
- Google Workspace blog: Generate longer AI videos with Omni 1.2 Flash in Google Vids at no cost (November 5, 2026)