Google’s most advanced video generation model, Veo 3, is now officially available in the Gemini API. It was first showcased at Google I/O 2025, and it comes with a high-fidelity video generation with native audio support to developers, researchers, and enterprises looking to build or prototype video experiences within the Gemini ecosystem.
With this release, developers can start using Veo 3 directly from Google AI Studio, where it’s available as an SDK template and a fully functional Starter App. Pricing starts at $0.75 per second of output for both video and audio, with a more affordable Veo 3 Fast variant on the way.
