Video Generations

Send a unified video request body as JSON. The model must be enabled for your workspace and routed to a channel that supports video task generation.

1curl -X POST "https://api.0-0.pro/v1/video/tasks" \2  -H "Authorization: Bearer $API_KEY" \3  -H "Content-Type: application/json" \4  -d '{5    "model": "seedance-2-0",6    "prompt": "A ginger cat sunbathes by a window, slow dolly-in, warm natural light, cinematic detail",7    "ratio": "16:9",8    "duration": 5,9    "resolution": "720p",10    "audio": false,11    "watermark": false,12    "seed": -113  }'

The create response returns a task id. Poll that id until the upstream response reaches a terminal state. A successful response usually includes a video URL and a usage object.

1curl "https://api.0-0.pro/v1/video/tasks/video-task-id" \2  -H "Authorization: Bearer $API_KEY"

1{2  "id": "video-task-id",3  "model": "seedance-2-0",4  "provider": "lingjing",5  "status": "succeeded",6  "video": {7    "url": "https://example.com/generated-video.mp4",8    "duration": 5,9    "resolution": "720p",10    "ratio": "16:9"11  },12  "usage": {13    "video_tokens": 108900,14    "completion_tokens": 108900,15    "total_tokens": 10890016  }17}

• 0-0 routes the unified request to a provider that supports the selected model and options.
• Provider-specific options can be sent in the provider_options object.
• When a poll response returns status: succeeded with usage tokens, 0-0 records the token usage once for billing and logs.
• If you query a task that was not created through this 0-0 deployment, add ?model=... so the gateway can route the poll request to the right upstream channel.