Google Gemini Omni 1.1 Flash Adds 4K AI Video and 40 Second Extensions
Google's Gemini Omni 1.1 Flash adds 40-second scene extensions, keyframe transitions, cheaper 360p previews, video references and 4K output for developers.
Summary
Google DeepMind product managers Anish Nangia and Alisa Fortin launched Gemini Omni 1.1 Flash on August 27, 2026, positioning the generative video model as production-ready through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform. It analyzes up to 10 seconds of prior footage, versus only the final second previously, and extends scenes in 10-second increments to a cumulative 40 seconds. Developers can specify first and last frames for continuous transitions, camera orbits, zooms and loops, or provide up to three seconds of reference video to preserve visual and character consistency.
New 360p previews generate up to 60% faster by system throughput and cost one-third as much as standard 720p output, supporting rapid prototypes and comparisons of three to four draft variations. Finished videos can be generated at 1080p or upscaled to 4K. Adobe has integrated Gemini Omni Flash into Firefly, while Figma Weave, GMI Cloud and Runway describe production uses spanning creative iteration, educational content and editing. Gemini Omni 1.1 Flash is rolling out globally in Google Flow for Google AI Plus, Pro and Ultra subscribers, with scene extension also available to those subscribers in the Gemini app.
Positives
- Scene extension now considers 10 seconds of context, improving continuity over models that referenced only the final second.
- 360p previews generate up to 60% faster and cost one-third as much as standard 720p output.
- First and last frame controls enable continuous camera movements, transitions and seamless loops.
- 1080p and 4K output targets professional production rather than draft-only workflows.
- Adobe Firefly, Figma Weave, GMI Cloud and Runway are already applying Gemini Omni Flash in production workflows.
Risks & concerns
- Scene extensions are limited to 10-second increments and a maximum cumulative length of 40 seconds.
- Reference video input is capped at three seconds, restricting how much source motion the model can follow.
- The 60% speed improvement applies specifically to system throughput for 360p compared with 720p.
- Google Flow and Gemini app access requires a Google AI Plus, Pro or Ultra subscription.