Generate videos using Google’s Veo models through the Gemini API.
Setup
export GOOGLE_API_KEY=your-key
Usage
from praisonaiagents import VideoAgent
agent = VideoAgent(llm="gemini/veo-3.0-generate-preview")
video = agent.start("A sunrise over mountains", output="sunrise.mp4")
Models
| Model | Description |
|---|
gemini/veo-2.0-generate-001 | Veo 2.0 |
gemini/veo-3.0-generate-preview | Veo 3.0 |
gemini/veo-3.0-fast-generate-preview | Veo 3.0 Fast |
gemini/veo-3.1-generate-preview | Veo 3.1 |
gemini/veo-3.1-fast-generate-preview | Veo 3.1 Fast |
Parameters
video = agent.generate(
prompt="Your description"
)
Gemini Veo has limited parameter control compared to OpenAI. Duration and size are handled automatically.
Features
| Feature | Supported |
|---|
| Text-to-Video | ✅ |
| Image-to-Video | ✅ |
| Video Remix | ❌ |
| Custom Duration | Limited |
