v1

latestOpenAPI 3.0.3MITraw.githubusercontent.com2025-02-09165953.5 KB

Analysis

Analyze - AI Vision Moderation

The Moderation mode accepts multiple questions about an image, to which the response provides concise answers of "yes," "no," or "unknown." This functionality allows for a nuanced evaluation of whether the image adheres to specific content policies, creative specs, or aesthetic criteria.

post/analyze/ai_vision_moderation

Request body

asyncboolean

Whether to perform the analysis asynchronously. If true, the response is returned immediately and analysis is done in the background. Once complete, a webhook notification will be sent to the specified URL and/or to the URLs defined in the Webhook Notifications settings of your Cloudinary Console.

notification_urlstring

The webhook URL to notify when the analysis is complete. Only relevant when async is set to true.

rejection_questionsstring[]

A list of yes/no questions

Example request

{
  "source": {
    "asset_id": "e12345b5c456c8901bbb0efc00c0fcf",
    "uri": "https://res.cloudinary.com/demo/image/upload/sample.jpg"
  },
  "notification_url": "https://path.to/webhook",
  "rejection_questions": [
    "Does the image contain any violent activity?"
  ]
}

Response

Analysis succeeded

request_idstring

Example response

{
  "data": {
    "analysis": {
      "responses": [
        {
          "prompt": "Does the image contain any violent activity?",
          "value": "no"
        },
        {
          "prompt": "Is there any nudity in the image?",
          "value": "yes"
        }
      ],
      "model_version": 1
    }
  },
  "limits": {
    "items": [
      {
        "type": "ai_vision",
        "used_by_request": 123,
        "remaining": 500,
        "limit": 10000,
        "reset_time": "2024-01-20T12:34:56Z"
      }
    ]
  },
  "request_id": "17c3b70c5096df0e77e838323abb7029"
}