Docs
Docs pages

API reference

POST /v1/identity Pending

One row per tracked animal: its track number in this clip, when it was seen, and an identity id when the animal is known from your gallery.

Add include_embeddings=true to get the embedding of each track. Track numbers are per clip: the same animal can get another number in the next clip. Matching animals across clips needs the identity capability, which has not passed its test yet.

Request

Send a video the same way as /v1/analyze (raw bytes, or video_b64 in JSON), with the same options: species, max_seconds, start_s, step_s. Or pass sample to get a stored result of a real run with no video: nothing is metered and the response says "sample": true.

curl -s -X POST "https://vitalsbio-console.pages.dev/api/v1/identity?sample=cow-lying" \
  -H "Authorization: Bearer $VITALSBIO_API_KEY"

curl -s "https://vitalsbio-console.pages.dev/api/v1/identity?species=cattle&max_seconds=20" \
  -H "Authorization: Bearer $VITALSBIO_API_KEY" -H "Content-Type: video/mp4" \
  --data-binary @barn.mp4

Samples: cow-lying, dog-walk, brangus-herd, paddock, white-horses. This endpoint runs the same model call as /v1/analyze and is metered the same way, by seconds of video. If you need more than one part of the same clip, call /v1/analyze once.

Response

FieldWhat it is
capabilityThe status of this part: live (passed its locked test), preview, pending or off. Read it before you rely on the numbers.
sampletrue when the answer is a stored result, not a run on your video.
animals[].track_idThe animal's number within this clip.
animals[].identity_idA name from your gallery of known animals. null until identity is Live and a gallery is set.
animals[].confidenceHow sure the match is, 0 to 1. null with no match.
animals[].first_seen_s, last_seen_s, n_framesWhen the track was seen, in seconds, and how many sampled frames it has.
animals[].embedding128 numbers describing how the animal looks. Only with include_embeddings=true. Compare two with cosine similarity.
usage.video_secondsSeconds of video metered for this call. Also in the X-Vitals-Video-Seconds header.
request_id, api_versionAlso in the X-Request-Id and Vitals-Version headers.

This is the real answer for sample=cow-lying, cut short:

{
  "object": "identity",
  "request_id": "req_Lr0d1sY2rE9q",
  "api_version": "2026-11-21",
  "schema_version": "biomodel-v0.1",
  "capability": {
    "id": "identity",
    "name": "Identity across days",
    "status": "pending",
    "mode": "preview",
    "label": "Pending"
  },
  "sample": true,
  "video": {
    "fps": 25,
    "width": 1920,
    "height": 1080,
    "n_frames": 689,
    "step_s": 0.2
  },
  "model": {
    "name": "vitals-biomodel",
    "version": "biomodel-v0.1"
  },
  "animals": [
    {
      "track_id": 1,
      "identity_id": null,
      "confidence": null,
      "species": "cattle",
      "first_seen_s": 0,
      "last_seen_s": 26.8,
      "n_frames": 135
    },
    {
      "track_id": 2,
      "identity_id": null,
      "confidence": null,
      "species": "cattle",
      "first_seen_s": 0,
      "last_seen_s": 4.8,
      "n_frames": 14
    }
  ],
  "usage": {
    "video_seconds": 0
  }
}

Errors, limits and headers are the same on every endpoint: see errors and rate limits. For a long video or a camera, see streaming.