Google has just introduced Veo 3, its latest video generation model, on the Gemini API. This tool allows developers to transform a simple text prompt into a high-resolution video complete with synchronized audio—capturing visuals, dialogue, music, and sound effects all in one go.
Right now, the API focuses on text-to-video conversion, with plans to extend its capabilities to image-to-video soon—the feature you may already have seen in the Gemini app. Developers can dive in using the SDK template and starter app available via Google AI Studio, although you’ll need an active Google Cloud project with billing enabled to get started. Early adopters have already been exploring its potential through the Gemini app, Flow, and Vertex AI.
The pricing model is straightforward: it costs $0.75 per second for 720p, 24fps video with audio in a 16:9 format—25 cents more per second than the earlier Veo 2, which didn’t include sound. There’s even talk of a ‘Veo 3 Fast’ mode that will offer quicker and cheaper access, though this isn’t available on the API just yet. If you’re planning longer projects, something like an eight-second clip comes to about $6, while a five‑minute video may cost around $225. And if you need multiple attempts to perfect your vision, keep in mind that costs can add up quickly.
Real-world applications are already in play. For example, Cartwheel is using Veo 3 to convert flat 2D videos into lifelike 3D animations by mapping generated movements onto rigged models. Similarly, game developer Volley is utilising the technology to craft cutscenes for its RPG ‘Wit’s End’, enabling rapid iterations on both narrative and visuals. It’s clear that while some projects remain under wraps, many industries are finding practical uses for this innovative model.
If you’ve ever wrestled with the challenges of creating production-level videos on a budget, Veo 3 offers a refreshing mix of technical finesse and user-friendly design that may just bridge the gap between your creative ideas and reality.
