Google introduces Flow, an advanced video generation tool

Google has introduced Flow, a new tool that will combine all of the company's generative models, such as Imagen for image creation, Veo for video generation, and Gemini for text-based creation of cinematic clips, scenes, and stories.

Users will be able to formulate a video idea for Gemini, create the necessary assets using Imagen, and then implement the idea through Veo's generative model.

According to Engadget, Flow will also have additional features for editing and staging videos. Users will be able to choose camera movements and angles, as well as edit and add new clips and entire scenes. In addition, Flow will have its own asset management system for organizing assets and even your cues.

Google's new tool will be available in the US with a Google AI Pro subscription or the newly introduced AI Ultra. In the case of AI Pro, users will get access to the Veo 2 model, which has received several improvements, such as the ability to control camera movements, a "Redraw" feature to expand the frame from portrait to landscape format, and a better understanding of shadows, scale, and the interaction of objects in the video.

At the same time, the new Veo 3 model will be available in the Google AI Ultra subscription. This is Google's most advanced video generation model, which, in addition to better quality compared to Veo 2, will also be able to generate videos with sound for the first time, such as traffic noise, birdsong, or even dialogue between characters.

We remind you that at Google I/O, the Gemini chatbot for Chrome was also introduced, the launch of the "Review by AI" extension in search in Ukraine and other regions, the Imagen 4 image generation model, and the Stitch UI creation tool were also introduced. In addition, Google announced that it is working with Xreal on Project Aura smart glasses based on Android XR.