Google has rolled out Veo 3.1, the upgraded version of its advanced Veo 3 video generation model, offering improved realism, refined textures, enhanced lighting, and cinematic-grade sound design. This update is part of a broader revamp of Flow, Google’s AI-powered filmmaking platform, which now features detailed scene-editing tools, options to add or remove objects, and the ability to produce longer and smoother video sequences.
Flow AI Receives a Significant Creative Boost
Initially introduced five months ago at Google I/O 2025, the Flow AI tool and Veo 3 model quickly gained attention from creators seeking deeper artistic and audio control. Responding to that feedback, Google has now upgraded the platform with Veo 3.1, enabling more flexible storytelling and enhanced creative freedom.
With this update, users gain improved control over both visuals and narrative flow while generating richer, more realistic audio. Additionally, the new Veo 3.1 model demonstrates stronger prompt accuracy and better alignment with the creator’s intended story direction.
You can now upload multiple reference images to define characters, objects, and the overall visual style. The Flow AI tool merges these images to build a unified, realistic scene. Additionally, the generated videos now come with integrated audio.
Flow also lets you input both a starting and ending image to automatically generate a smooth transition video connecting the two. You can even extend an existing clip to create longer sequences — up to a minute or more — making it ideal for producing continuous shots. The tool uses the final frame of your video to naturally carry the motion forward.
Moreover, Flow allows you to insert new elements or objects into any scene, automatically adjusting lighting and shadows for enhanced realism. You can also erase unwanted items or characters effortlessly. The latest Veo 3.1 model is now available for use in both the Gemini app and Flow starting today.
