Video
GA
Proprietary
Veo 3.1
Cinematic quality with audio and camera control
Release2026-04
Context100K
Max output—
ParametersUndisclosed
Local / VRAMAPI only
LicenseProprietary
Overview
Veo 3.1 is a model in the Video category from Google DeepMind. It was released in 2026-04. The vendor is based in US/UK. Best for: Cinematic quality with audio and camera control.
Strengths
- 4K
- Native dialogue/audio
- 3 reference images
- Scene extension
- Strong physics
Limits
- 8 sec clips
- 4K is expensive
- Closed
Specifications
| Vendor | Google DeepMind |
|---|---|
| Category | Video |
| Release | 2026-04 |
| Status | GA |
| License | Proprietary |
| Commercial use | Not specified |
| Context window | 100K |
| Input modalities | Text, Image, Video |
| Output modalities | Video, Audio |
| Prompt types | Text, Image, Reference |
| Negative prompts | No |
| Max outputs | 1-4 clips |
| Resolution | 720p, 1080p, 4K |
| Duration | 4/6/8s |
| Frame rate | 24 fps |
| Audio generation | Yes |
| Local deployment | No |
| API access | Yes |
| Pricing | $0.05-0.60 / second |
| Confidence | High |
| Vendor country | US/UK |
| Vendor type | Frontier lab |
Pricing & hardware
$0.05-0.60 / second
API-only: local GPU memory is not applicable. API access is available.