Vision & Document AI APIs
Eight capabilities across identity, vision, and document intelligence — each with a live demo you can try right now. No account required.
Production-ready APIs
Face Similarity
Available1:1 biometric identity matching
Match a selfie to an ID or compare two faces. Returns a match score, confidence, and decision verdict for KYC and re-verification flows.
Face Recognition
AvailableSearch a face against your enrolled gallery
Identify a person by searching one face against millions of enrolled identities. Returns ranked candidate matches with similarity scores.
Face Detection & Analysis
AvailableDetect faces and extract rich attributes
Detect every face in an image and return bounding boxes, landmarks, age, and gender estimates — all in a single API call.
Face Blur
AvailableAnonymize faces for privacy compliance
Automatically detect and blur faces for GDPR compliance, media redaction, and privacy-first data pipelines. Batch API supported.
Face Liveness
AvailableAnti-spoofing and presentation attack detection
Passive liveness check for onboarding flows. Detects print, replay, and 3D mask attacks without requiring user interaction.
Beta — early access
OCR
BetaExtract text from any document
High-accuracy text extraction powered by Claude. Handles receipts, invoices, handwriting, and multi-language documents.
Document AI
BetaStructured data from IDs and forms
Extract typed, schema-validated fields from passports, driver's licenses, invoices, and contracts with per-field confidence.
Document Classification
BetaClassify documents before routing
Classify an uploaded document by type — passport, driver's license, invoice, bank statement, and more — with a confidence score, as a triage step before OCR or Document AI.