Google · Clips · premium
Veo 3.1
Google's flagship: generates dialogue and sound from the same prompt as the picture.
What it costs
| 5s · silent | 83.5 credits |
| 5s · with audio | 167 credits |
What it takes
| Aspect ratios | 16:9 9:16 |
| Resolutions | 720p 1080p |
| Duration | 4, 6, 8 seconds |
| Reference images | 3 |
| Last frame | Accepted |
When to reach for it
The model to use when a line of dialogue is the point. Veo writes its own audio from the same prompt that produces the image: speech, room tone, footsteps, sometimes score. Quoting a line verbatim in the prompt gets you that line rather than a mouth moving around approximate syllables.
Sound is a price switch, not a post-process. A five-second take is 83.5 credits silent and 167 with audio. That is worth it for a spoken line and rarely worth it for a landscape, so decide per shot rather than leaving it on.
Three reference images, the tightest of the models that take any, plus a last-frame slot, which is the real mechanism for continuity: render a shot, take its final frame, open the next take with it. Two aspect ratios only, 16:9 and 9:16, so a square or 4:5 deliverable has to be cropped. The Fast and Lite variants drop references entirely, which is the thing to check before planning a sequence around them.
Made with it

A creator who does not exist
He holds the camera. He does the talking.

The sound comes with it
Eight seconds, audio made in the same pass.
Compared with
- Veo 3.1 vs Seedance 2.5
Veo 3.1 when someone speaks. Seedance 2.5 when the shot is long, or has a cast to keep consistent.
- Veo 3.1 vs Veo 3.1 Fast
Fast for a shot that stands alone. Full Veo the moment the clip has to match something else.
Guides that cover it
- Two pictures, fifteen seconds: the first-and-last-frame method
Make the opening frame and the closing frame yourself, hand a video model both, and let it fill the middle. Nine models take a last frame. What it buys.
- Veo 3.1: the complete prompting guide
Eight seconds of video that writes its own soundtrack. Three audio channels, eight rules, four pipelines and the failures worth knowing about first.
Veo 3.1 is one pick in the composer.
Every model in this catalog runs from the same prompt bar, priced in the same credits. New accounts start with some to spend.
Start picturing