Fuel your creativity with new generative media models and tools

Details

Google DeepMind launched two new generative media models, Veo 3 and Imagen 4, alongside a filmmaking tool called Flow. The most significant technical capability is Veo 3's ability to generate video with synchronized audio — generating sound effects, ambient noise, and dialogue alongside the visual content, which previous generative video models could not do natively. Imagen 4 targets higher fidelity image generation with improved text rendering and photorealism. Both models were developed in consultation with filmmakers, photographers, and visual artists — signaling a strategic emphasis on professional rather than just consumer use cases. Flow is a dedicated filmmaking tool that provides scene-level control over narrative structure, camera angles, and character consistency across shots. This addresses one of the main practical barriers to AI video in professional production: the lack of continuity controls that allow editors to build coherent visual sequences. Access to Veo 3 and Flow is initially limited to Google One Ultra subscribers and enterprise users, suggesting premium positioning. For creative professionals and production studios evaluating AI tools, these launches represent a meaningful upgrade over the current state of the generative video market.