Google has made its Gemini Omni 1.2 Flash video model generally available, extending native video generation to 15 seconds and cutting the price of 720p generation by 30 percent. Google DeepMind announced the release on 5 November, and the model is now live through the Gemini API and Gemini for business.

The update matters because video models have long been limited to short clips that creators then had to stitch together. Gemini Omni 1.2 Flash generates synchronised video and native audio in a single pass, for any whole-second duration from three to 15 seconds, up from a 10-second ceiling in the previous version, Gemini Omni 1.1 Flash.

Longer clips and a lower price

Google prices standard 720p output at $0.07 per second, 30 percent lower than Omni 1.1, so a full 15-second clip at 720p costs about $1.05. Developers call the model as gemini-omni-1.2-flash on the paid Gemini API tier. Google says the model is also rolling out through Flow, Google Vids, the Gemini app for Google AI Ultra subscribers, and Google Ads, although access conditions vary by product and plan.

In Google Vids, eligible users with a Google account or Workspace subscription receive an allowance of three free 15-second videos a month. Adobe also plans to add Gemini Omni Flash 1.2 to Firefly, where creators already have access to the earlier Gemini Omni model.

Conversational editing across a whole clip

The release is not only about length. Gemini Omni 1.2 Flash supports conversational editing across full 15-second clips, so a creator can ask for changes in plain language and keep the rest of the scene intact, rather than regenerating from scratch. Clips can also be extended by three to 10 seconds per turn, up to 30 seconds in total, and the model offers stronger reference following, holding on to a supplied image or style across edits.

Google says every video the model generates carries a SynthID watermark that is imperceptible to viewers but detectable by software, and that safety filters apply to prompts and outputs. The company is open about the model's limits, noting that developers are still working on fast sports action, rapid script changes and complex physics.

What happens to the old model

Gemini Omni 1.1 Flash remains available for a four-week transition. From 3 December 2026, Google says existing 1.1 API requests will automatically route to 1.2 with no code change, and will be billed at the lower 1.2 rates. Creators should treat that as a production migration, not only a billing change, and check that their workflows still behave as expected after the switch.

There is one regional caveat. Editing or extending uploaded video is unavailable in the European Economic Area, Switzerland and the United Kingdom, although Google says editing and extension of model-generated video remain supported in available regions.

For developers and small studios, the combination of longer clips, cheaper 720p output and in-place editing makes AI video generation a more practical tool for short advertisements, social clips and storyboards. The test now is whether the quality holds up over a full 15 seconds, and whether creators adopt it widely enough for Google's price cut to matter.