Google Releases Gemini Omni 1.2 Flash With 15-Second Video Clips and a 30% Price Cut

Google on November 5, 2026 made Gemini Omni 1.2 Flash, its video generation model, generally available in the Gemini API and Google Cloud. The model makes clips of up to 15 seconds, up from 10 on Omni 1.1, and Google cut the price by 30% to about $0.07 per second of 720p video, the company said.

Omni is a young product line. Google introduced the Gemini Omni family at its I/O conference on May 19, 2026 and opened Omni Flash to developers on June 30 at $0.10 per second, with clips capped at 10 seconds (The Next Web). Omni also anchors how scenes look in Genie 4, the world model among Google's October model releases.

What did Google release?

Gemini Omni 1.2 Flash is an update to Google DeepMind's video model that adds 15-second clips, editing over several chat turns and closer adherence to reference images, Google said in its announcement on November 5, 2026. Users can set the length in one second steps from 3 to 15 seconds.

The model takes character, product or style images, motion clips of up to 3 seconds and audio cues as references, and Google says faces, clothing, packaging text and brand colors now stay consistent across camera angles. Conversational editing lets a user change one element of a finished clip, such as the background, lighting or aspect ratio, without generating it again. Google also claims more natural dialogue with clearer lip sync, and camera moves with fewer unwanted jump cuts.

Google lists its own limits: work continues on fast sports action, rapid script changes and complex physics.

How much does Gemini Omni 1.2 Flash cost?

Gemini Omni 1.2 Flash costs $12 per million video output tokens in the Gemini API, which Google says works out to about $0.07 per second of 720p video, down from about $0.10 on Omni 1.1, according to Google's pricing page as of November 2026. The model is only on the paid tier.

Google also suggests a cheaper workflow: draft at 360p, fix details with follow-up edits, then upscale the best clip to 1080p or 4K. Set side by side, the pricing page and the announcement show what the longer clip costs: a full 15-second Omni 1.2 clip at 720p comes to about $1.04, while a 10-second clip, the old maximum, cost about $1.01 on Omni 1.1 by the same token rate. That is a calculation from Google's listed rates, not a figure Google states, and it means 50% more video for roughly the same spend.

Older code will not stay on the old model for long. From December 3, 2026, all API requests to Omni 1.1 Flash are routed to Omni 1.2 Flash and billed at 1.2 rates with no code changes required, Google's documentation says.

How does Omni 1.2 compare with Omni 1.1 and rival models?

In Google DeepMind's own side-by-side tests, raters preferred Omni 1.2 Flash over Omni 1.1 Flash on 8 of 8 rows, with win rates from 54% to 62%, according to charts on the Gemini Omni model page. The widest gap is visual consistency in text to video, at 62% against 38%; the narrowest are edit instruction adherence and text to video visual quality, both at 54%.

Those charts compare Omni 1.2 only with its predecessor. Google published no results against rival video models, and no independent benchmark has been released yet.

Clip length is where rivals still lead. ByteDance's Seedance 2.5, announced on June 23, 2026, makes 30-second clips in native 4K from a single prompt (The Next Web). Omni 1.2 reaches 30 seconds only through extension: Google's developer documentation says a clip can be extended by 3 to 10 seconds per turn, up to 30 seconds in total, and its 4K output is upscaled rather than native. The documentation also limits clips of 11 to 15 seconds to one reference video, against up to three for shorter clips.

Who can use Gemini Omni 1.2 Flash, and where?

Developers can use Gemini Omni 1.2 Flash in the Gemini API and Google AI Studio, and businesses through Google Cloud, where Google says Gemini Omni is generally available for the first time, backed by Google Cloud's infrastructure and security. Provisioned Throughput, which reserves capacity at a fixed rate, is also offered.

For consumers, Google says the model is rolling out in Flow, Google's AI filmmaking app, in Google Vids, where 15-second clips come at no extra cost with a Google account or Workspace subscription, and in the Gemini app for Google AI Ultra subscribers. Advertisers get it in Google Ads. The model is also listed on OpenRouter, a service that offers many AI models through one API. Every output carries a SynthID watermark, Google says.

What we don't know yet

• How Omni 1.2 Flash scores against Seedance 2.5, Veo and other video models in independent tests.
• How much conversational editing saves in practice, since Google has published no figures on the cost of a typical edit turn.
• Whether all features and consumer rollouts reach the European Union at the same time as the United States.
• When the Gemini app rollout to Google AI Ultra subscribers will be complete.