Google has released Gemini Omni 1.1 Flash for video generation

Google has released Gemini Omni 1.1 Flash — an updated video generation model ready for use in production development. The model is available via the Gemini API in Google AI Studio, and enterprise customers can work with it through the Gemini Enterprise Agent Platform.

The main innovation is the model's ability to analyze up to 10 seconds of preceding video when extending a scene, whereas previous versions relied only on the last frame. This helps maintain characters, environments, and the overall direction of the scene more consistently. Videos can be extended in 10-second segments up to a total length of 40 seconds.

Developers can also specify the first and last frames of the video — the model will generate the transition between them: camera movement around an object, smooth zooming, or a looping clip. For additional visual context, up to three seconds of existing video can be provided to the model — for example, to have it reproduce a character in a newly generated scene.

A draft testing mode is available for experimenting with ideas, generating videos at 360p resolution — it runs 60% faster and is three times cheaper compared to the standard 720p resolution. The final video can be obtained in 1080p or scaled up to 4K.

Omni 1.1 Flash is also available to users with Google AI Plus, Pro, and Ultra subscriptions in the Google Flow app — starting today. The scene extension feature will also appear in the main Gemini app.

Source: blog.google