Google released Gemini Omni 1.2 Flash on November 5, the latest version of its Omni video model. Clips can now run up to 15 seconds, up from 10 in Omni 1.1, and Google says the price is 30% lower, at $0.07 per second for 720p video. The model is generally available through the Gemini API and Google Cloud, and Google is also putting it into Flow, Google Vids, the Gemini app and Google Ads.

What changed from Omni 1.1

The headline change is length. Google says Omni 1.2 Flash generates and edits clips natively up to 15 seconds, and developers can set the duration in one-second steps from 3 to 15 seconds. A longer single shot matters because stitching several short clips together is where characters and objects tend to drift.

Google lists several quality changes, all described qualitatively:

  • Reference adherence. The model accepts character, product or style images, short motion clips of up to 3 seconds, and audio cues. Google says faces, garments, packaging typography and brand colors stay consistent across camera angles.
  • Consistency. Characters, objects and scenes should hold from the first frame to the last with less drift or morphing. First and last keyframe interpolation and clip extension are supported.
  • Speech. Dialogue is described as more natural, with clearer lip-sync and distinct conditioning for several speakers.
  • Camera control. Continuous moves such as first-person flythroughs, zoom-outs and orbits are supported, with fewer unwanted jump cuts.

Google’s post gives no numeric benchmark results for these claims, so they are the company’s own assessment and we have not been able to verify them independently.

Editing in turns instead of regenerating

The workflow Google describes is draft first, polish later. Users generate a draft at 360p or 720p, fix details through follow-up instructions, and only then upscale the final clip to 1080p or 4K. Follow-up edits can swap a product or background, change lighting or weather, or change the aspect ratio across the full 15 seconds. The practical point is cost control: a cheap low-resolution draft is regenerated or edited until it is right, and the expensive upscale happens once.

Outputs carry an imperceptible SynthID watermark, Google’s provenance marker for AI-generated media.

Where you can use it

According to Google, access depends on the product:

  • Gemini API and Google Cloud: generally available, including Vertex AI Studio, an enterprise API and Provisioned Throughput for reserved capacity.
  • Flow: multi-turn directing and editing.
  • Google Vids: 15-second clips at no additional cost with a Google account or a Workspace subscription.
  • Gemini app: 15-second videos, rolling out to Google AI Ultra subscribers.
  • Google Ads: support for scaling video campaigns.
A person working at a laptop in an office
Generic office scene: Google Vids brings 15-second Omni clips to everyday office accounts. Photo: Bench Accounting benchaccounting / Wikimedia Commons, CC0

How it compares on price and length

Against its own predecessor the numbers are clear: 15 seconds instead of 10, and a price 30% lower. At the quoted $0.07 per second, a full 15-second 720p clip works out to about $1.05 before any upscaling, a figure we calculated from Google’s per-second rate. That is useful for planning ad variants or product videos at volume, where dozens of drafts may be needed per final clip.

Comparison with rival video models is harder. Competitors price and cap clip length differently, and Google published no head-to-head scores. Our earlier coverage of Kling 4.0 and its stereo audio shows how quickly rivals are adding sound and speed-focused variants, and our explainer on how AI video generators work covers why temporal consistency is the hard part.

What it means for marketers and small studios

Product-video teams gain the most from reference-driven generation: a packaging shot or a model photo can anchor a clip, and edits can be applied without starting over. Google lists a 15-second virtual try-on sequence built from one model photo and front and back shots of three outfits as an example. Because Vids includes 15-second clips at no extra cost for existing accounts, many office users can try the feature without a separate video budget.

Limits remain. Google says it is still working on fast sports action, rapid script changes and complex physics, which are the cases most likely to expose artifacts.

What to watch next

  • Independent comparisons of Omni 1.2 Flash against other video models on consistency and lip-sync.
  • Whether Google extends the 15-second cap or brings the same controls to a higher-tier Omni model.
  • How the Gemini app rollout to Google AI Ultra subscribers progresses.
  • Any change to the per-second price for 1080p and 4K upscaling.

Key takeaways

  • Gemini Omni 1.2 Flash was released on November 5 and is generally available via the Gemini API and Google Cloud.
  • Clips reach 15 seconds, and 720p generation costs $0.07 per second, 30% less than Omni 1.1.
  • Multi-turn editing lets users refine a draft instead of regenerating it, with upscaling to 1080p or 4K at the end.
  • Google’s quality claims come with no published benchmark numbers.
  • Outputs carry a SynthID watermark.

Frequently asked questions

How long can Gemini Omni 1.2 Flash clips be?

Up to 15 seconds, set in one-second steps from 3 to 15. Omni 1.1 was limited to 10 seconds.

How much does Gemini Omni 1.2 Flash cost?

Google lists $0.07 per second for 720p video, which it says is 30% lower than Omni 1.1. Pricing for upscaling to 1080p or 4K was not stated in the post we reviewed.

Is the video watermarked?

Yes. Google says outputs carry an imperceptible SynthID watermark.