Veo 3.1 is Google’s cinematic shot generation model, featuring professional-grade resolution, native audio, and reference-driven characters, scenes, and styles. Prompt your shot from nothing, animate a still, or adjust an existing video with words.
Everything Veo 3.1 can do
Sound born with the shot
Dialogue, foley and score arrive in the same pass as the pixels, matched to what is on screen rather than laid over it afterwards. Turn Generate Audio off for a silent plate, which is also the cheaper render.
Three tiers on one picker
Fast for iteration, Veo 3.1 for the hero shot, Image to Video for animating a still. A tier the current settings rule out stays visible with the reason attached instead of vanishing from the list.
Animate a still
Attach one image and give it motion at 720p or 1080p, with the camera move and the ambience written into the same prompt.
Keep the shot rolling
Attach a clip, choose how many seconds to add, and the extension continues the same scene, camera language and sound — up to 148 seconds all told.
Frequently Asked Questions
How long can a Veo 3.1 clip be?
4, 6 or 8 seconds at 720p. Above 720p, or with reference images attached, the clip is always 8 seconds.
Does it generate sound?
Yes. Dialogue, foley and score are produced in the same pass as the video on every Veo 3.1 route, extensions included. Turn Generate Audio off for a silent render, which costs less.
What is the difference between Fast, Veo 3.1 and Image to Video?
Fast is the cheap iteration tier and Veo 3.1 the full-quality one; both take a text prompt and both reach 4K. Image to Video runs on Veo 3.1 Lite, needs one attached still to animate, and tops out at 1080p.
How do reference images work?
Attach up to three stills to guide a text render. The route requires an 8-second clip whenever references are attached, and they are ignored while you are extending a clip.
What does extending a clip do?
Attach a clip, pick how many seconds to add, and the render continues from its last frame with matching motion and sound. Input plus extension can reach 148 seconds.
Which aspect ratios and resolutions are supported?
16:9 or 9:16, at 720p, 1080p or 4K. 4K is on the text routes only; animating a still tops out at 1080p.
Can I control whether people appear?
Yes. The People setting sends the render a policy — allow everyone, adults only, or nobody at all.
