Google introduced Gemini Omni 1.1 Flash on August 27, 2026, positioning the update as production-ready generative video for developers who need more control than a one-shot clip. The model is available through the Gemini API in Google AI Studio and on the Gemini Enterprise Agent Platform.

Scene extension is the headline control. Omni 1.1 can analyze up to 10 seconds of prior context, up from models that only looked at the final second, then continue a clip in 10-second increments to a cumulative 40 seconds. Developers can also specify first and last frames so the model interpolates a continuous shot, add up to three seconds of video as a reference for character consistency, draft at 360p, and upscale finished output to 1080p or 4K.

Who is already using it

Google says 360p drafts generate up to 60% faster and at a third of the cost of Omni 1.1’s standard 720p mode, which is aimed at storyboard iteration. Adobe has integrated Gemini Omni Flash into Firefly. Figma Weave, GMI Cloud, and Runway are named as customers using the model in production creative tools.

The same controls are rolling out to Google AI Plus, Pro, and Ultra subscribers in Google Flow. Scene extension is also in the Gemini app for those tiers.

Decoded Take

Video models have been easy to demo and hard to direct. Omni 1.1 is Google trying to sell control (start and end frames, cheaper drafts, 4K finish) as the difference between a toy and an editing tool. Putting it in Firefly, Figma, and Runway matters more than the 40-second cap, because those hosts already have users who think in shots rather than prompts. The constraint to watch is consistency across an extended scene: if the extra context window still drifts characters or lighting after two or three continuations, professional teams will keep Omni for boards and finish elsewhere. Measure that in customer case studies, not launch reels.