Quickstart

Make your first call with a test key: compare two faces with face.verify. You need a VeriSec account and about five minutes.

1. Create a test key

   On verisec.dev, open API keys and create a key. Test keys start with sk_test_. The secret is shown once, so store it straight away, in an environment variable rather than in your code:

   export VERISEC_API_KEY=sk_test_…

2. Install a library

   Pick your language, or skip this step and use cURL.

   # cURL needs nothing extra.
   curl --version

3. Create a client

   Every library reads VERISEC_API_KEY from the environment when you don't pass a key.

   # cURL sends the key on every request:
   #   -H "Authorization: Bearer $VERISEC_API_KEY"

4. Compare two faces

   face.verify answers whether two photos show the same person. The libraries take a file path for each image and send it for you; with cURL you send the image as base64.

   curl -X POST https://verisec.dev/v1/face/verify \
     -H "Authorization: Bearer $VERISEC_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "image1": "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNkYPhfDwAChwGA60e6kgAAAABJRU5ErkJggg==",
       "image2": "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNkYPhfDwAChwGA60e6kgAAAABJRU5ErkJggg==",
       "threshold": 0.75
     }'

5. Read the answer

   KYC answers share one envelope. data is the result; request_id is what you quote if you contact us.

   {
     "ok": true,
     "request_id": "req_cd1d47ea5bf34b7aae6c",
     "latency_ms": 2362,
     "data": {
       "code": 0,
       "msg": "OK",
       "similarity": 0.9983906745910645,
       "score": 0.23992152512073517,
       "match": true,
       "threshold": 0.75,
       "compared": [
         {
           "image": "image1",
           "roi": [
             182,
             241,
             173,
             220
           ]
         },
         {
           "image": "image2",
           "roi": [
             940,
             245,
             176,
             216
           ]
         }
       ],
       "elapsedTime": 2.299846035,
       "metadata": {
         "serverTime": "2026-10-05T05:35:01.160Z",
         "ctx": {
           "userIp": "203.0.113.7"
         }
       }
     },
     "warnings": [],
     "error": null
   }

   match is the decision at the threshold you sent (0.75 when you leave it out), and similarity is the score behind it.

Next steps#