GoogleVideo modelLifelike, with a voice.
Skin, fabric, light and weight behave the way a camera records them. When the brief says it must look real, this is the model people reach for.
Spokesperson ads
A presenter who says your line on camera, in a setting that reads as real.
Short scenes with dialogue
Two characters, one exchange, eight seconds. The building block of a narrative ad.
Premium product shots
Hero moments in 4K where realism carries the price point.
Vertical social
9:16 at the same quality as widescreen, for Reels, TikTok and Shorts.
- Provider
- Resolution
- 720p, 1080p, 4K
- Duration
- 4, 6 or 8 seconds
- Aspect ratios
- 16:9, 9:16
- Audio
- Dialogue, effects and ambience, on or off
- Modes
- Text to video, image to video
- Frames
- Start and end frames
- Seed
- Fixed seed for repeatable takes
- 8s with sound
- 534 credits at 720p or 1080p, 800 at 4K
- Plans
- Starter and up
- Cost
- From 267 credits per 8 seconds
Veo 3.1 Fast
Same look for 200 credits per 8 seconds with sound, when speed and budget matter.
Kling 3.0
You need longer than 8 seconds, or a square frame.
FLUX 3 Video
FLUX 3 VideoYou want up to 20 seconds of Full HD with sound for less.
Seedance 2.0
Seedance 2.0You want to steer the shot with reference images, clips and audio.
Veo 3.1 Lite
Veo's motion, drafted.
Side by side with the models people weigh it against. Both run on the same plan.
Video: Kling AIVeo 3.1 vs Kling 3.0
Google's realism against Kuaishou's cinema. One stops at 8 seconds, the other runs to 15.
Read the comparisonSeedance 2.0 vs Veo 3.1
ByteDance's reference-led control against Google's realism. Both reach 4K.
Read the comparisonFLUX 3 Video vs Veo 3.1
Black Forest Labs' value flagship against Google's. Sound included on one, 4K on the other.
Read the comparisonWhat is Veo 3.1?
Veo 3.1 is Google DeepMind's flagship video model. It generates 4, 6 or 8 second clips at 720p, 1080p or 4K, in 16:9 or 9:16, from a text prompt or a start and end frame, and renders spoken dialogue, sound effects and ambience with the picture. Its strength is realism: faces, skin, light and physical behaviour read as filmed rather than generated, and lines of dialogue land with matching lips. On APImage an 8 second clip costs 267 credits silent or 534 with sound at 720p or 1080p, and 800 with sound at 4K. Veo 3.1 Fast and Veo 3.1 Lite are the cheaper ways to draft the same shot.
How much does Veo 3.1 cost on APImage?
An 8 second clip is 267 credits silent or 534 with sound at 720p or 1080p, and 534 silent or 800 with sound at 4K. Shorter clips cost proportionally less: 4 seconds with sound is 267 at 1080p.
How long can a Veo 3.1 clip be?
4, 6 or 8 seconds per generation. To go longer, Veo 3.1 Fast Extend adds 7 seconds to a clip in Veo quality.
Does Veo 3.1 generate audio?
Yes. Dialogue, effects and ambience are rendered with the picture. Sound is a toggle, and switching it off halves the price.
Veo 3.1, Fast or Lite?
Lite to draft cheaply, from 40 credits per 8 seconds. Fast for most finished work at about a third of Veo 3.1's price with sound. Veo 3.1 when realism is the whole brief.
Which plan includes Veo 3.1?
Starter and up. It draws on the same credits as every other model, with no separate Google subscription.
How do I use it?
Open the video studio and pick Veo 3.1. Write the shot, including the exact line anyone speaks and what we hear. Add a start frame, and an end frame if the move has a destination. Pick 16:9 or 9:16, the resolution and the length, then generate.