Image Content Moderation API — POST /v1/moderations

POST /v1/moderations with model gi-image-moderation checks images for NSFW content across 5 categories including neutral, sexy, and explicit.

The /v1/moderations endpoint also accepts images for NSFW and content policy classification. Pass an image URL or a base64-encoded data URI in your request to receive per-category confidence scores and a top-level flagged boolean — giving you everything you need to gate image uploads or flag content for review.

Request

POST https://kiosapi.com/v1/moderations

Headers

Content-Type · string · required

Must be application/json.

Authorization · string · required

Your API key in the format Bearer sk-xxx.

Request Body

model · string · required

Must be gi-image-moderation for image classification.

input · array · required

An array of image objects to moderate. Each object describes one image.

Show input[] fields

input[].type · string · required

Must be "image_url".

input[].image_url.url · string · required

The URL of the image to moderate, or a base64-encoded data URI (e.g. data:image/png;base64,...).

Example Request

cURL

curl https://kiosapi.com/v1/moderations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "gi-image-moderation",
    "input": [
      {
        "type": "image_url",
        "image_url": {
          "url": "https://example.com/image.png"
        }
      }
    ]
  }'

Python

import openai

openai.api_key = "sk-xxx"
openai.base_url = "https://kiosapi.com/v1/"

response = openai.moderations.create(
    model="gi-image-moderation",
    input=[{
        "type": "image_url",
        "image_url": {"url": "https://example.com/image.png"}
    }]
)

print(response.results[0].flagged)
print(response.results[0].categories)

Response Categories

The image moderation model scores each image across five categories:

Category	Description
neutral	No sexual content detected
drawings	Normal image with slight sexual possibility (e.g. illustrated or animated content)
sexy	Provocative or suggestive content
hentai	Adult anime or illustrated sexual content
porn	Explicit sexual content

Scores across categories are independent confidence values — they do not sum to 1. Evaluate each category score against your own threshold to suit your application’s moderation policy.

Response

{
  "id": "abfc99e9-392e-44e3-aedd-e63f33079491",
  "model": "gi-image-moderation",
  "results": [
    {
      "flagged": false,
      "categories": {
        "neutral": false,
        "drawings": true,
        "sexy": false,
        "hentai": false,
        "porn": false
      },
      "category_scores": {
        "neutral": 0.2273,
        "drawings": 0.9944,
        "sexy": 0.1233,
        "hentai": 0.2199,
        "porn": 0.0557
      }
    }
  ]
}

Response Fields

id · string

A unique identifier for this moderation request.

model · string

The classifier model that processed the image.

results · array

An array of moderation result objects, one per image submitted.

Show results[] fields

results[].flagged · boolean

true if the image triggered one or more content categories; false otherwise.

results[].categories · object

A map of each of the five category names (neutral, drawings, sexy, hentai, porn) to a boolean indicating whether that category was triggered.

results[].category_scores · object

A map of each category name to a floating-point confidence score. Use these scores to implement custom severity thresholds tailored to your platform’s content policy.