Google launches new video tool Flow alongside its latest model Veo 3 with dialogue support

Google has introduced a new 'AI filmmaking tool' called Google Flow, allowing users to generate short videos from text or image prompts. Users can generate 8-second clips and connect them with scenebuilder tools to build longer sequences. Input options include plain text or a mix of text and reference images, a feature Google names 'ingredients-to-video.' In a demonstration video, the company showed how users can create multiple 8-second clips with different camera angles, scenarios, or models, and combine them in Flow to form a sequence with smooth transitions and consistent visual style. There’s also a 'Flow TV' page with a gallery of AI-generated video examples.

Alongside Flow, Google also announced Veo 3, its latest AI video generation model offering improved video quality, support for longer prompts, and notably, synchronous audio generation with included dialogue. This represents a major step forward over earlier models, as audio and dialogue synchronization is now possible. Google's existing Veo 2 model also received camera control and object removal updates.

Google also launched Imagen 4 for high-fidelity image generation and improved text rendering. Flow is currently available only to U.S. users through the Google AI Pro plan, which includes core Flow features and 100 generations per month, or the new $250/month AI Ultra plan, which increases usage limits and provides early access to Veo 3 with native audio support.