Google had several exciting AI announcements at I/O 2025, including new Project Astra features that will be available in Gemini Live in the near future, a new AI Mode for Google Search that will change the way you shop, and hardware demos of the first AR/AI smart glasses it plans to ship to consumers later this year. But there’s no denying that Veo 3 (along with Flow) was one of the most exciting things Google unveiled this year.
Veo 3 is Google’s next-gen AI video generation software, and it’s a massive upgrade over its predecessor. Veo 3 can produce ambient noise, sound, and even dialogue for your AI-generated movies. All you need to do is enter a text prompt that includes all the details, and Veo 3 will generate the desired results.
Veo 3 will sync spoken language to the clip and will let you preserve characters from one scene to the next. Veo 3 went viral online after I/O 2025, and we saw a few amazing videos and short movies created with it. Also, an actual movie playing in theaters will incorporate shots made with Veo 3.
The biggest problem with Veo 3 is that it’s expensive to access. You need a $249.99/month Google AI Ultra subscription to make the most of Veo 3. You can also gain access to Veo 3 from various third-party services, including Adobe’s Firefly, Freepik, and LTX Studio. But you’ll still have to pay a subscription fee in order to use Veo 3.
A more cost-effective alternative, coming June 27, is to go to Google’s all-in-one AI cloud tool Vertex AI and use Veo 3 in public preview. This will let you try Veo 3 before you determine whether you need this AI video generator for personal use or work.
Google said in a blog post that all Google Cloud customers and partners can access Veo 3 in public preview on Vertex AI. According to the company, Veo 3 brings these three features to AI video generation in Vertex AI:
Fluid, natural videos that synchronize video with audio and dialogue. Veo Renato can synchronize your audio and visuals in a single pass. The model produces rich soundscapes containing everything from dialogue and ambient noise, to sound effects and background music.
Cinematic video that captures creative nuances. Veo 3 makes it easy to capture creative nuances and detailed scene interactions in your prompt, from the shade of the sky to the precise way the sun hits water in the afternoon light, and produces high-definition video.
Realistic movement that simulates real-world physics. To create believable scenes, Veo 3 simulates real-world physics. This results in realistic water movement, accurate shadows connected with objects and characters, and natural human motion.
Google also posted a few Veo 3 samples, including the short clip it first showed at I/O 2025.
Google shared additional Veo 3 clips made via third-party tools like Freepik and LTX Studio, alongside comments from these companies, which are working with Google Cloud to create AI videos for various purposes:
Here’s a sample made via the Freepik AI Video Generator:
This one was made via LTX Studio:
Finally, the next clip is a new brand created with Gemini, Imagen, and Veo 3:
To try Veo 3 in Vertex AI, you can get started at this link. All you need is a Google Cloud account.
