Google announced Veo, its first AI video generation model, in May 2024. Generative video was one of the key AI-related fields the company had yet to enter, so it was only natural that Google would take the plunge at some point. Now, the firm has announced the Veo 3 model alongside Flow, a brand-new AI filmmaking tool.
Google’s first steps in AI generative video were decent but could be better. In December, the company announced Veo 2 as an improved version that promised much more natural results. Now, during the Google I/O 2025 event we met Veo 3. The latest update to the AI model promises to be even better, but Google doesn’t want it to be limited to just another simple video generation tool.
Veo 3-powered Flow will make things much easier for creatives
Flow has the potential to change the way storytellers and creative minds bring their cinematic visions to life. The AI filmmaking tool is specifically built to work with Google’s most advanced AI models. That is, it not only uses Veo, but also the Imagen and Gemini LLMs.
Imagine a creative process where ideas simply flow, where time seems to slow down, and you can effortlessly bring scenes to life. That’s the inspiration behind Flow. Regarding Veo 3, you can expect amazing visuals that really stick to your prompts and look incredibly realistic, even when it comes to physics.
What makes it even smoother? Gemini models are working behind the scenes to make prompting super intuitive. You can just describe your vision in everyday language, and Flow will get to work. Plus, you’re not limited to just text; you can bring in your own characters and assets. You can even create new elements using Imagen’s text-to-image capabilities. Once you’ve made a subject or a scene, you can reuse those elements consistently across different clips. This is a huge time-saver for maintaining continuity between scenes.
Multiple options available
Flow comes packed with features to help your storytelling truly shine. Camera Controls allows you to take direct command over camera motion, angles, and perspectives to get the perfect shot. Scenebuilder can extend your existing shots, transition between moments, and keep characters consistent. Asset Management keeps all your creative ingredients and prompts neatly organized. Flow TV brings together clips and content generated with Veo where you can even see the exact prompts and techniques used, which is fantastic for learning and adapting new styles.
Flow is an evolution of VideoFX, an earlier Google Labs experiment. If you’re eager to try it out, Flow is available starting today for subscribers of Google AI Pro and Google AI Ultra plans in the US. Google AI Pro gives you 100 generations per month, while Google AI Ultra offers higher usage and native audio generation. In other words, the “Ultra” tier allows you to add environmental sounds and character dialogue to your videos.
It’s early days for Flow, and Google is keen to shape its future with feedback from creatives. The tool has the potential to be a game-changer for digital storytelling, but only time will tell if it succeeds.
