Google video generation

Veo 3.1

Google's flagship video model with native audio, first/last-frame control, and subject reference images.

Examples

See what Veo 3.1 can make

What it is good for

Generate 720p or 1080p video with synchronized audio, controlled first and last frames, or up to three subject reference images.

Explore camera language, atmosphere, pacing, and sound before production.

Keep a key subject recognizable with reference-guided generation.

Create a planned movement between two supplied compositions.

Key capabilities

Create dialogue, ambience, and effects together with the visuals for a more coherent result.

Guide a shot from a chosen opening composition to a specific ending frame.

Use up to three references to preserve a subject across an eight-second landscape clip.

How to get a strong result

Start with the subject, then add lighting, style, and anything that must stay the same. Upload a reference whenever you want a more controlled edit.

1
Write sound into the prompt
Describe dialogue, ambience, and effects alongside the visual action.
2
Separate subject and camera motion
State what moves in the scene, then specify the camera move and lens behavior.
3
Respect reference mode
Reference images automatically use 16:9 and eight seconds; the last-frame input is ignored.

Series compare

Choose flagship control, faster iteration, or cost-efficient volume.

Feature	Veo 3.1Current	Veo 3.1 Fast	Veo 3.1 Lite	Veo 3	Veo 3 Fast
Best for	Premium cinematic work	Fast creative iteration	Scaled content creation	Cinematic generation	Rapid iteration
Speed	Standard	Fast	Optimized	Standard	Fast
Native audio	Configurable	Configurable	Always on	Configurable	Configurable
Frame control	First + last	First + last	First + last	First frame	First frame
Subject references	1–3 images	—	—	—	—
Default resolution	1080p	1080p	720p	1080p	1080p