Detect
Query inference results from the leading models on the BitMind Subnet.
The detect model
The detect model contains all the information about the inference results.
Properties
- Name
isAI
- Type
- boolean
- Description
Whether the content is predicted to be AI generated or not.
- Name
confidence
- Type
- float
- Description
Confidence score of the inference result.
- Name
predictions
- Type
- array
- Description
Predictions of the inference result.
- Name
similarity
- Type
- float
- Description
Similarity to known AI generated content.
- Name
fqdn
- Type
- string
- Description
Fully qualified domain name of the validator that generated the inference result.
Submit an image
This endpoint allows you to submit an image for detection.
When submitting images, you can choose between providing a URL or Base64 encoded image data. For URLs, use this option when your image is already hosted and publicly accessible or when you want to avoid the overhead of encoding images. For Base64 encoding, use this when your image isn't publicly hosted or when you need to process client-side images.
Required attributes
- Name
image
- Type
- string
- Description
URL or Base64 encoded image.
When using a URL, the image must be publicly accessible.
When using a Base64 encoded image, the max size is 4MB. Supported MIME types:
image/gif
,image/jpeg
,image/png
,image/bmp
,image/tiff
.
Request
curl -X POST https://subnet-api.bitmindlabs.ai/detect-image \
-H "Authorization: Bearer {token}" \
-H "Content-Type: application/json" \
-d '{
"image": "https://example.com/image.png"
}'
Response
{
"isAI": true,
"confidence": 0.9991202464509517,
"predictions": [
0.9567890123456789,
0.9992146492004395,
// ...
],
"similarity": 0.26494999999999996,
"fqdn": "validator.bitmindlabs.ai"
}