
Detect deepfakes and face manipulations using BYNN SOTA model
EFFORT deepfake detection for images and videos. Trained on DF40 dataset with 31 deepfake methods. AUC: 99.924%
image_urlstringURL of the image to analyze
https://example.com/face.jpgbase64_imagestringBase64-encoded image data
/9j/4AAQSkZJRgABAQAA...video_urlstringURL of the video to analyze (extracts 8 frames)
https://example.com/video.mp4base64_videostringBase64-encoded video data
AAAAIGZ0eXBpc29t...Deepfake detection results with confidence scores
is_fakebooleanTrue if deepfake detected
trueis_realbooleanTrue if authentic/real content
falsefake_probabilityfloatProbability that content is fake (rescaled: 0.9-1.0 becomes 0-1)
0.85confidencefloatModel confidence score
0.9998labelstringClassification label
fake{
"model": "effort-deepfake",
"image_url": "https://example.com/face.jpg"
}{
"success": true,
"data": {
"is_fake": true,
"is_real": false,
"fake_probability": 0.85,
"confidence": 0.9998,
"label": "fake"
}
}429 HTTP error code along with an error message. You should then retry with an exponential back-off strategy, meaning that you should retry after 4 seconds, then 8 seconds, then 16 seconds, etc.Integrate Video Deepfake Detection into your application today with our easy-to-use API.