by Google
Google's video generation model producing high-fidelity clips with cinematic camera control.
Veo 2 by Google DeepMind is a state-of-the-art video generation model that turns text and images into 8-second video clips with accurate real-world physics, diverse cinematic styles, and extensive camera controls. Known for its strong understanding of cinematographic language and natural scene dynamics.
Veo works best with short, powerful prompts focused on the essential visual elements:
Drone shot over autumn vineyards at golden hour, rows of
grapevines stretching to misty mountains, warm color palette
Include specific camera and lens terminology:
Wide 18mm lens, dolly in through a foggy redwood forest,
shallow depth of field, dappled sunlight, golden hour
Describe how the scene unfolds — not how it looks as a still image:
A flock of starlings swirling in a murmuration against
a pink sunset sky, camera tracks upward following the pattern
| Parameter | Description |
|---|---|
| aspect_ratio | 16:9, 9:16, 1:1 |
| duration | Clip length in seconds |
| resolution | Up to 4K |
| fps | Frame rate |
| seed | Reproducibility seed |
Quick tips from the community about what works with Veo 2 right now.
Sign in to share a tip.
No tips yet. Add a tip for this model.