Gemini Omni Flash 1.1 extends videos to 40 seconds

An opening frame, a closing frame: Gemini Omni Flash 1.1 from Google is designed to generate camera pans, zoom transitions and seamless loops in between. The new version also extends existing scenes with 10 seconds of video context instead of the previous 1 second, and can extend videos in ten-second increments up to a total of 40 seconds.

According to market observers, the key difference from a conventional video model lies in its architecture. Omni is therefore not a foundational video model that creates a new video from a single prompt, but an orchestration layer. Google describes the experience more as an ongoing conversation with a director: existing material, start and end frames, and short references set the direction.

To speed up selection, Google is lowering the barrier for previews. 360p videos are expected to be generated 60 percent faster and made available at one-third the cost of the 720p standard resolution. Users satisfied with the result can then upscale it to 1080p or 4K. It is also possible to provide three seconds of video reference as visual context.

In practical terms, this changes the workflow for users of Google Flow and the available APIs: they can first review scenes with cheaper 360p previews, control transitions using opening and closing frames, and only then request a higher resolution. Omni 1.1 Flash is available through the Gemini API in Google AI Studio and through the Gemini Enterprise Agent Platform API; in Flow, it is open worldwide to subscribers of Google AI Plus, Pro and Ultra.

The assessment nevertheless remains measured. Omni Flash competes with Seedance 2.0 and Kling 3.0, but has so far been seen as trailing their capabilities. Google’s figures for speed, cost and the new features come from the company; the information available does not mention an independent comparison of version 1.1.