AI image & deepfake detection

Add a review signal for AI-generated and manipulated images, with a three-state verdict and likely generator attribution.

POST/ai-detectionmultipart/form-data

Request

Send a JPEG or PNG in the image multipart form-data field. The published maximum file size is 20 MB.

Form field	Type	Required
`image`	Image file	Yes

Set RAPIDAPI_KEY and RAPIDAPI_HOST to the values from this API’s listing. The following example uses a POSIX shell. Replace the local filenames with your images.

cURL

curl --request POST \
  --url "https://$RAPIDAPI_HOST/ai-detection" \
  --header "X-RapidAPI-Key: $RAPIDAPI_KEY" \
  --header "X-RapidAPI-Host: $RAPIDAPI_HOST" \
  --form 'image=@./image.jpg'

Response example

JSON

{
  "verdict": "LIKELY_AI_GENERATED",
  "aiGeneratedScore": 0.999,
  "deepfakeScore": 0.02,
  "sources": [
    {
      "name": "stablediffusionxl",
      "score": 0.991
    }
  ]
}

Sample response excerpt with illustrative values.

Core response fields

This guide focuses on the fields used in application workflows. Responses may include additional metadata; consult the API listing for the full schema.

Field	Type	Meaning
`verdict`	string	LIKELY_AI_GENERATED, UNCERTAIN, or LIKELY_AUTHENTIC.
`aiGeneratedScore`	number	AI-generation risk signal from 0 to 1.
`deepfakeScore`	number	Deepfake risk signal from 0 to 1.
`sources[]`	array	Likely generators, each with a name and score, most-confident first.

Interpreting the result

Use the verdict and scores to prioritize image review. Scores are model signals rather than calibrated probabilities; image quality, compression, and generation methods affect interpretation. Include a review path for uncertain results.

Limits and integration notes

The published image upload limit is 20 MB. Resize files that exceed the limit before sending them.
Keep keys server-side and use the correct host for the API. See authentication.
Handle bad inputs and temporary failures using the error guide.
Vision input images are processed in-request without storage by Tigrate. See data handling.

Try AI image & deepfake detection on RapidAPI

Based on the published RapidAPI reference, checked September 2026. Consult the listing for the latest full schema, optional parameters, and plan limits.

Questions about this integration?Contact the team