Google's Gemini AI Video Generator Gets Major Upgrade with New Controls

Google's Gemini AI video generator has been upgraded with new features that give creators more control over their generated videos. The latest update, called Omni 1.1 Flash, allows developers to extend existing footage and generate motion between specified frames.

The model can now extend an existing scene in 10-second increments, up to a cumulative 40 seconds, using up to 10 seconds of previous footage as context for the next generation.

This means that creators can continue footage they already like rather than regenerating the entire sequence. They also have more control over how a generated sequence transitions between two points by providing a first and last frame.

However, the model still has clear limits. It can only generate 3-10 seconds of video in a single pass, so longer sequences require repeated extensions. The model's 1080p and 4K options also rely on upscaling rather than native generation at those resolutions.

The update is available across Google's developer and creator tools, including Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform. The new capabilities are also available in Google Flow.