Veo 3 API Usage

Hello,

I’m currently developing an app that utilizes Veo 3 via the Gemini API, and I was wondering what the easiest way to implement Google Flow’s “jump to” feature is. I’ve already implemented an extend feature by extracting the last frame of the video, but I am struggling to figure out a way to replicate jump to. Would it just be a matter of getting the selected frame and prompting (e.g., “While maintaining consistency, jump to the following scene: {prompt}”), or would there be a better way to do this?

Let me know your thoughts on this. Any potential implementations would be greatly appreciated. Thanks.

Hello,

Welcome to the Forum!

Currently, Veo models accept text and images as input. You can try providing some of the previous frames as input and then prompt the model to maintain continuity. Unfortunately, there’s no jump to or extend feature available in Gemini API.