DOCUMENT TO JSON

Turn documents into structured JSON

Define a schema, extract its fields, and keep uncertainty visible for review rules.

INPUT invoice.pdf + schema OUTPUT { "total": 1843.50 }

{
  "total": 1843.50,
  "currency": "USD",
  "supplier": "Acme Co"
}

WHAT TO EVALUATE

JSON that matches your schema

Required, nullable, nested and enum fields

Nested fields

Know what needs review

Nullable values

Avoid different response formats

Field confidence

Read fields in the right context

INTEGRATION

Use one request in development and production.

curl -X POST https://api.startocr.com/v1/extract \ -H "Authorization: Bearer $STARTOCR_API_KEY" \ -F file=@invoice.pdf \ -F schema=@invoice.json

POST /v1/extract

curl -X POST https://api.startocr.com/v1/extract \
  -H "Authorization: Bearer $STARTOCR_API_KEY" \
  -F file=@invoice.pdf \
  -F schema=@invoice.json

COMMON QUESTION

Which output should I use?

Use structured extraction when a system needs named fields and a stable response contract rather than raw text.

USAGE BASED

Structured extraction from $20 / 1K pages

Start with representative documents, then pay only for the pages processed.