Google has introduced a new tool for video creation using generative AI — “Flow“. It allows users to create video clips up to eight seconds long based on text prompts or a combination of text with their own images. With the “Scenebuilder” feature, multiple clips can be combined into a full scene, and the “Camera Controls” tool allows for adjusting the movement and angle of the camera in the video.
The tool operates on the new Veo 3 model, which now generates video along with synchronized audio — from character dialogues to ambient sounds. This allows for creating more realistic content, considering the details of motion physics and accurate speech reproduction. Users also gain access to the Imagen 4 model, which has improved image quality, supports various formats, and better handles text reproduction on images.
“Flow” offers additional features such as “Asset Management” for organizing materials, tools for adding and removing objects, and “outpainting” for changing framing and expanding the scene. Additionally, users can take advantage of “Flow TV” — a showcase of video examples using prompts and techniques, helping to quickly master the new application.
The tool is already available in the US for subscribers of the Google AI Pro and Google AI Ultra plans. The first plan provides one hundred video generations per month, while the second offers extended limits and early access to the new Veo 3 model with audio support. Google notes that in developing “Flow,” it collaborated with professional filmmakers to make the interface convenient for both experienced users and beginners.
