Content moderation

Process image, video, text, face, and voice risks in one place.

Input mode

Images, GIFs, and videos use URI access for online assets, uploaded files, and replay tasks.

Risk dimensions

Cover erotic, violent, political, and child safety risks, with reserved text, face, and voice evidence.

Structured result

Return levels, labels, age groups, and video-frame evidence for policies, review, and reporting.

Integration mode

Support sync scans and async callbacks for real-time publishing, batch review, and training validation.

Media input
No media selectedUpload media to preview it and show real scan results.
Risk types

Choose the risk dimensions for this scan. If none are selected, the API scans every type by default.

Scan result

Waiting for scan results

Request example
{
  "is_video": false,
  "risk_types": [
    "EROTIC",
    "VIOLENT",
    "CHILD"
  ],
  "uri": "https://example.com/image.jpg"
}
Response example
{}