LogoOmniVideo API
Back to OmniVideo API

Quickstart

Create your first Gemini Omni video task

This guide uses one generate request and one status request. Keep the API key on your server.

1

Create an API key

Sign in and create a Bearer token from the developer dashboard.

Open API Keys
2

Submit a generation task

curl -X POST https://omnivideoapi.com/api/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-omni/video",
    "input": {
      "prompt": "A cinematic drone shot over a glass city at sunrise",
      "mode": "std",
      "aspect_ratio": "16:9",
      "duration": "5",
      "sound": true
    }
  }'

The response includes a task id.

{
  "code": 200,
  "message": "success",
  "data": {
    "task_id": "task_abc123",
    "status": "IN_PROGRESS"
  }
}
3

Poll status

curl "https://omnivideoapi.com/api/status?task_id=task_abc123" \
  -H "Authorization: Bearer YOUR_API_KEY"

Poll every 5 to 10 seconds. When the task succeeds, read the generated video URLs from the response payload.

Image-to-video variant

Add public image URLs to the input object to guide the generation.

{
  "model": "gemini-omni/video",
  "input": {
    "prompt": "Camera slowly pushes in, soft studio light",
    "image_urls": ["https://example.com/product.jpg"],
    "mode": "pro",
    "aspect_ratio": "1:1",
    "duration": "5",
    "sound": false
  }
}