Google Gemini Omni 1.2 Flash: 15-second video at $0.07 a second

Google made Gemini Omni 1.2 Flash generally available on 5 November 2026, according to a post by Anish Nangia, a product manager at Google DeepMind. The video-generation model now makes clips up to 15 seconds long and costs 30% less than Gemini Omni 1.1, at $0.07 per second for 720p video, Google says.

What changed

• Length: native generation and editing up to 15 seconds, set in one-second steps from 3 to 15 seconds. Omni 1.1 stopped at 10 seconds.
• Price: Google says it dropped the price 30% from Omni 1.1 to $0.07 per second at 720p. Google's post does not state the old price; the cut implies roughly $0.10 per second, which is my arithmetic, not Google's figure.
• Editing: multi-turn conversational editing, which lets a team swap a product or background, change lighting or weather, or switch aspect ratio on the same clip. Google says this works across full 15-second clips.
• References: Google says the model can combine character, product or style images, motion clips of up to three seconds and audio cues, and keep faces, garments, packaging typography and brand colours consistent across camera angles.

What it costs to use

At $0.07 per second, a 15-second 720p clip works out to about $1.05 (15 × $0.07). That is the generation price only. Google's post gives no price for 1080p or 4K and points readers to its pricing page, so the cost of the final upscaled version is not stated in the sources I read.

Google describes a tiered workflow to keep spend down: draft at 360p or 720p, fix a detail with a follow-up edit rather than regenerating the whole clip, then upscale the best result to 1080p or 4K. Whether that saves money depends on how many edit turns a clip needs, and each turn's price is not broken out in the announcement.

Where it runs and what it leaves out

Google lists the Gemini API and Google AI Studio, the Google Cloud console, Flow, Google Vids, Google Ads and, rolling out, the Gemini app for Google AI Ultra subscribers. Google says 15-second clips in Google Vids come at no additional cost with a Google account or Google Workspace subscription. Provisioned Throughput, which reserves capacity, is available, and every output carries an imperceptible SynthID watermark that marks it as AI-generated.

Google's own caveat is that development is ongoing on fast sports action, rapid script changes and complex physics. The post cites example prompts, not benchmark scores, and The Decoder's coverage likewise relays Google's claims. I found no independent evaluation, so treat consistency and lip-sync improvements as the vendor's description until you test them on your own footage.

What to do with it

Teams already generating product videos or ad variants should rerun a costed sample: the same brief at 720p, 15 seconds, with two edit turns, then compare against your Omni 1.1 invoice. Brand teams should check that logo and packaging text hold across camera angles, since Google highlights that as an improvement but publishes no measurement.