POST an image, get back the 17-character VIN and a confidence score. Windshield etchings, dashboard plates, door jamb stickers, registration documents — and a companion endpoint that does the same for license plates.
REQUEST
curl -X POST \
"https://api.carapi.dev/v1/extract-vin\
?token=YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"image":"<base64-jpeg>"}'RESPONSE
{
"vin": "WBAKU210X00R62021",
"confidence": 0.97
}A 200 with "vin": null means no VIN was found in the image — that is an answer, not an error, so there is nothing to retry.
Let an inspector or seller photograph the windshield instead of typing 17 characters — fewer typos, faster onboarding.
Pull the VIN straight off a registration document or insurance scan and push it into the rest of the pipeline.
Read plates from vehicle photos with region-specific format validation across 27 countries.
Both OCR endpoints accept a base64-encoded image up to 15 MB; downscaling phone photos to roughly 1080p keeps requests small without hurting accuracy. Feed the extracted VIN straight into VIN Decode for specifications, or the plate into Plate to VIN to resolve it to a vehicle.
Free tier with 100 requests/month — enough to test OCR against your own photos.
Get Started Free