Google has released Gemini Omni 1.1 Flash as a production-ready video model for developers, adding controls aimed at turning short generated clips into more deliberate sequences. The model can extend scenes for up to 40 seconds, interpolate between a specified first and last frame, and upscale finished work to 4K.
A new 360p preview mode is designed for quicker, cheaper iteration before creators commit to higher-resolution output. Google says the update also improves visual consistency and narrative flow when extending a scene, addressing a common problem in generated video where subjects or environments change unexpectedly between shots.
Developers can access the model through Google AI Studio and the Gemini Enterprise Agent Platform. The controls do not remove the need to review generated footage for continuity and accuracy, but they give production teams clearer ways to set a shot’s boundaries and reserve expensive rendering for the final version.