On November 5, 2026, Google announced on its official blog that Gemini Omni 1.2 Flash, its video generation model, is now generally available across the Gemini API and Google Cloud. The release stretches a single clip from 10 to 15 seconds, adds multi-turn conversational editing, holds reference images more faithfully, and cuts 720p pricing by 30% to $0.07 per second. For teams producing video at scale, the update attacks time and cost at once.
Length is the most visible change. Omni 1.1 topped out at 10-second clips; 1.2 Flash generates natively up to 15 seconds and lets you specify any duration from 3 to 15 seconds in one-second steps, so one ad or product explainer can fit in a single generation. The workflow is cheaper too: draft fast at 360p or 720p, fix a detail with a follow-up conversational instruction instead of re-rendering the whole clip, then upscale the best take to 1080p or 4K. On reference fidelity, the model can take in character, product, or style images, motion clips up to three seconds, and audio cues, keeping faces, garments, packaging typography, and brand colors consistent throughout. Google's example is virtual try-on: one model photo plus front-and-back shots of three outfits produces a 15-second, 360-degree try-on sequence, with each garment's cut and back details intact.
Advertisers and e-commerce teams benefit first: the same asset can change background, lighting, or aspect ratio through conversation, so variants for different audiences no longer need a fresh brief. Google Vids and the Gemini app support 15-second generation too, with Gemini app access rolling out to Google AI Ultra subscribers. Dialogue and lip-sync are more natural, and multi-speaker scenes assign voices per character. Outputs still carry an imperceptible SynthID watermark, and enterprise teams can reserve throughput with Provisioned Throughput. Google is candid about the limits: fast sports action, frequent script changes, and complex physics are still being improved, and 15 seconds is still not long-form video.
Developers can try it now in Google AI Studio, where the documentation and pricing pages are already updated; businesses can connect through the Google Cloud console. Existing Omni 1.1 workflows need no rewrite — switch the model to 1.2 Flash and the new pricing applies.
