Unified Detect

One endpoint that auto-detects the media type and routes detection.

A single endpoint that inspects your input, determines whether it's an image, video, or text, and runs the right detector — so you don't have to call a type-specific endpoint. This is the recommended front door for new integrations.

Endpoint

POST https://api.bitmind.ai/v1/detect

Authentication

Send a Bearer token in the Authorization header.

See Authentication.

How routing works

The media type is resolved automatically:

• Binary uploads (file, base64, or a downloaded URL) are identified by their content — JPEG/PNG/etc. → image, MP4/WebM/etc. → video.

• Plain text is detected implicitly — no flag required. A text/plain body, a JSON text field, or a JSON media string that isn't a URL/base64 medium is scored as text.

• Bytes that match no supported media format are rejected with 415.

You may optionally set type to force routing (image, video, or text). A type that contradicts the actual bytes returns 400.

Request

Auto-detected URL

curl -X POST https://api.bitmind.ai/v1/detect \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"media": "https://example.com/photo.jpg"}'

Multipart file upload

curl -X POST https://api.bitmind.ai/v1/detect \
  -H "Authorization: Bearer $API_KEY" \
  -F "file=@clip.mp4"

Plain text (implicit)

curl -X POST https://api.bitmind.ai/v1/detect \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: text/plain" \
  --data "In conclusion, the synergistic leveraging of cross-functional..."

JSON fields

Field	Type	Required	Description
media	string	No*	URL or base64 data URI of an image/video
text	string	No*	Text passage to analyze (alternative to media)
type	string	No	Force routing: image, video, or text (otherwise auto-detected)
debug	boolean	No	Include debug data in the response
postContext	object	No	Source context, forwarded to text detection

* Provide exactly one of media, text, or a multipart file.

For video, the trim parameters startTime, endTime, and fps are honored, just like Detect Video.

Response

The response is an envelope: mediaType tells you what was detected, and result is the same payload the type-specific endpoint returns.

{
  "mediaType": "image",
  "result": {
    "isAI": false,
    "confidence": 0.23,
    "similarity": 0.05,
    "objectKey": "1234567890.jpg"
  }
}

Field	Type	Description
mediaType	string	Detected type: image, video, or text
result	object	The native detection result for that type

See Detect Image, Detect Video, and Detect Text for the per-type result shapes.