Google’s Veo 3 video-generation AI model, introduced in late May, is coming to developers.
“Starting today, we’re bringing Veo 3 to developers in paid preview via the Gemini API and Vertex AI,” Google product managers said in a blog post Thursday.
“Veo 3 will be priced at $0.75 per second for video and audio output,” they wrote. “Additionally, Veo 3 Fast will be available soon, offering a faster and more cost-effective option for video creation.”
Since Veo 3 was unveiled at the annual Google I/O conference in Mountain View, Calif., tens of millions of high-quality videos have been generated with Google’s first video model to incorporate high-fidelity video outputs and native audio, as well as text-to-video.
Google cited two recent examples: Cartwheel, which created a system that can translate two-dimensional videos of humans into fully-production-ready 3-D animation on rigged characters, and Volley, whose designers produced in-game video-cut scenes for an upcoming role-playing game called Wit’s End.
The inclusion of Veo 3 in the Gemini API and Google AI Studio essentially puts Google’s cutting-edge video generation directly into the hands of developers, enabling creators and developers to generate high-quality videos with synchronized dialogue, music, and sound effects from a single prompt.
When Google showed off Veo 3 nearly two months ago, it anticipated a wave of hyper-realistic eight-second clips stitched together into ads, imagined film trailers, and humorous street interviews. Academy Award–nominated director Darren Aronofsky used the tool to create a short film called “Ancestra.”
At the time, Google DeepMind CEO Demis Hassabis compared the advanced features of Veo 3 to “emerging from the silent era of video generation.”
Indeed, the age of video-generated AI content is in full swing: Google’s announcement arrived the same day that Adobe Inc. announced Firefly video updates, including the ability to add sound effects to Firefly AI videos.
