You define the schema
Send the fields and types you want back. The engine extracts only those — predictable, typed JSON every time.
Define the fields you care about, send us a PDF, and OCR Engine returns clean, structured JSON. No templates to maintain, no manual review.
Built for teams automating invoices, receipts, forms, contracts & reports.
A single endpoint that understands layout, language and intent — so you ship faster.
Send the fields and types you want back. The engine extracts only those — predictable, typed JSON every time.
Tables, multi-column pages and scanned documents are handled natively — no brittle coordinate templates.
Sub-3-second responses for most documents, with webhooks and batch endpoints for high volume.
Encrypted in transit and at rest. Files are processed and purged — never used for training.
Multilingual and mixed-script documents are recognised out of the box, including handwriting.
A single REST endpoint and an API key. Drop it into any stack in minutes — examples included.
Request access through the form below. We send a key you can start testing with right away.
List the fields and types you want extracted in a single JSON object — no training required.
POST your PDF and get back exactly the parameters you defined, ready to store or process.
There's nothing to configure server-side. Describe the data you want in the request body and OCR Engine maps the document to your schema — returning a confidence score alongside every field.
string, number, date, array, object# Define exactly what you want back
curl https://api.ocrengine.io/v1/extract \
-H "Authorization: Bearer YOUR_API_KEY" \
-F "file=@statement.pdf" \
-F 'fields={
"account_no": "string",
"balance": "number",
"transactions": "array"
}'
Tell us a little about what you're building and we'll get you set up with a key and onboarding details. No credit card required to start.