Google gives Gemini Omni finer control over AI video

Google has released Gemini Omni 1.1 Flash, a production-oriented update to its generative video model for developers using Google AI Studio and the Gemini API. The company says the update is aimed at creative tools, media editing software, and video workflows that need more control than one-shot text-to-video prompts.

What changed #

The biggest addition is scene extension. Omni 1.1 Flash can look at up to 10 seconds of prior video context and continue the scene in 10-second increments, up to 40 seconds total. Google says that longer context is meant to improve visual consistency and narrative continuity.

The model also adds first-and-last-frame control, so builders can provide the opening and closing images for a shot and let the model generate the transition between them. For teams iterating on prompts, Google added 360p previews before final rendering, then 4K upscaling for polished outputs.

Why it matters #

AI video tools are moving from impressive short clips toward workflows that look more like production software: controllable takes, edits, transitions, and cheaper drafts. Omni 1.1 Flash does not solve every reliability problem in generated video, but it gives developers more handles for building repeatable creative systems instead of isolated demos.