summary
Google's VEO 3 video generation model is now available through the Gemini API. This is the price range that makes it one of the more expensive options for AI video.
Gemini API Integration targets developers who are bringing advanced video generation into their own apps or building production-ready prototypes. At the moment, the API is restricted to text to video, but image to video support is on the way in which people live in Gemini apps. VEO 3 is Google's first model capable of generating high-resolution video and synchronous audio from a single text prompt. Create visuals, dialogs, music, and sound effects at once.
https://www.youtube.com/watch?v=3nxt1kxl1ny
To help developers get started, Google AI Studio offers a starter app to provide SDK templates and starter apps. Access requires an active Google Cloud Project with billing enabled. According to Google, VEO 3 has already been used millions of times in the Gemini app, Flow and Vertex AI.
Video with audio $0.75 per second
VEO 3 access via the Gemini API is only available at the paid tier of Google Cloud. Prices are $0.75 per second at 720p, 25 cents more than the Veo 2, which has 24fps video with 16:9 format audio – soundless. Google has also announced a faster and cheaper “VEO 3 FAST” mode, but is not available in the API yet.
advertisement
At the current rate, an 8-second video costs $6, while a 5-minute video costs $225. Producing perfect results can often involve multiple trials, which can lead to rapid rise in costs. For example, if you need 10x more footage to finish with a 5-minute available video, the total cost could reach $2,250. Still, Google may bet that in some use cases this may be cheaper than traditional video production.
Real World Examples
According to Google, Cartwheel uses VEO 3 to turn 2D videos into realistic 3D character animations, mapping generated movements to rigged models in client projects.
https://www.youtube.com/watch?v=wi8j_eqbl1y
Game Studio Volley uses VEO 3 to create cutscenes for the role-playing game Wit's End, allowing developers to quickly experiment with new story ideas and visuals. So far, these examples point to rather specialized use cases, and may suggest that Google has not yet emphasized any major integrations. It is also possible that some companies use VEO 3 behind the scenes, but are not ready to be published.
