Google makes Gemini Omni 1.2 Flash generally available with 15-second video

On 5 November 2026, Google DeepMind announced Gemini Omni 1.2 Flash and made it generally available on the Gemini API and Gemini for business. The update stretches native video generation and editing from 10 to 15 seconds, adds multi-turn conversational edits, and cuts the list price by 30% versus Omni 1.1, to $0.07 per second for 720p video.

The same day, Google Cloud confirmed general availability with enterprise controls such as Provisioned Throughput. Outputs carry an imperceptible SynthID watermark.

What changed from Omni 1.1

According to the DeepMind product post and the Google Cloud blog:

• Length: generate and edit clips up to 15 seconds, up from 10. Duration can be set in one-second steps from 3 to 15 seconds.
• Price: $0.07 per second at 720p, 30% below Omni 1.1 Flash, with similar savings at other resolutions.
• Draft then upscale: create drafts at 360p or 720p, then upscale selected clips to 1080p or 4K.
• References: combine character, product or style images, short motion clips (up to three seconds), and audio cues. Google says faces, garments, packaging type and brand colours stay more consistent across camera angles. Audio references remain allowlist-gated.
• Conversational edits: refine the same clip turn by turn — swap a product or background, change lighting, or switch aspect ratio — instead of regenerating from scratch, across the full 15 seconds.
• Speech: Google claims more natural dialogue, clearer lip-sync and distinct multi-speaker conditioning.

Google also lists open work: fast sports action, rapid script changes and complex physics still need improvement (DeepMind).

Where you can use it

Availability spans developer and consumer surfaces (DeepMind):

• Developers: Gemini API via Google AI Studio and Google Cloud.
• Business: Gemini for business and the Cloud console, with API embedding for apps and agents.
• Google apps: Flow for directed scenes and multi-turn edits; Google Ads for video campaigns; the Gemini app for Google AI Ultra subscribers (rolling out); Google Vids for Workspace and consumer accounts.

On Vids specifically, Google says anyone with a Google account or Workspace subscription can generate up to three 15-second videos per month at no extra cost.

Built for volume ads and product video

The Cloud post frames Omni 1.2 Flash for everyday business video: product clips, virtual try-ons and ads (Google Cloud). Retailers can turn existing stills into short product spins and lifestyle clips. Creative platforms quoted in that post — including Artlist, FLORA, invideo, Krea, Picsart, Photoroom, QuickFrame AI and Runway — say they are wiring the model into production tools.

A few claims from those customers, as published by Google Cloud:

• Photoroom’s CEO says 15-second video was a top user request for its Video Ad Maker and Product Animator, and that GA with SLAs lets them run it in production for sellers (Google Cloud).
• QuickFrame AI’s CTO says 15-second generations at 1080p let users create a complete connected-TV ad in one pass (Google Cloud).
• invideo’s VP of engineering says the model “sticks to the prompt” and that 15 seconds covers a full scene in one go (Google Cloud).

Those are vendor quotes republished by Google, not independent benchmarks.

Why it matters for AI wait time and ads

Fifteen seconds is a common unit for social and CTV spots. Lower per-second pricing plus draft-at-720p-then-upscale lowers the cost of iterating variants — the same pattern advertisers already use when they test a new channel with a small budget. For teams that already sit inside Google Vids, Flow or Ads, the model arrives where the creative work already happens, with a small free monthly allotment on Vids.

If you are comparing generative video APIs for catalog or wait-screen creatives, the concrete numbers from Google are: 15 s max, $0.07/s at 720p, 30% cheaper than 1.1, GA on Gemini API and Cloud, and three free 15-second Vids clips per month. Everything else — consistency, speech quality, reference fidelity — should be checked against your own product photos and brand guidelines before you scale spend.