Structured data from documents, processed in the EU.

An API that turns invoices, contracts and forms into typed JSON — built so that a compliance review is a short conversation instead of a six-week blocker.

curl -X POST https://api.sharpocr.com/v1/parse \
  -H "Authorization: Bearer $SHARPOCR_KEY" \
  -F file=@invoice.pdf \
  -F schema=eu_invoice

{
  "supplier":    { "value": "Acme B.V.",   "confidence": 0.99 },
  "vat_id":      { "value": "NL812345678B01", "confidence": 0.98,
                   "checksum_valid": true },
  "total_gross": { "value": 1210.00, "currency": "EUR", "confidence": 0.97 },
  "line_items":  [ /* … */ ],
  "checks":      { "line_items_sum_matches_total": true }
}

Illustrative response. The API is not open yet — this page exists to find the people who need it.

Why another OCR API

Short sub-processor list

An EU data centre for compute, one EU-incorporated provider for the hard pages. The full chain gets published before launch — including each provider's own sub-processors, which is where most “EU-only” claims quietly fall apart.

Confidence you can act on

Per-field confidence on every value, plus validation: line items sum to the total, VAT IDs checksum, dates make sense.

Your schemas

Define the fields and types you want back. Get those, or an explicit typed error — not a wall of raw text to post-process.

Delete-by-default

Documents deleted after processing. Zero-retention mode for anything you cannot let touch disk.

Indicative pricing

Plan	Monthly	Included
Free	€0	Evaluation volume, no card
Starter	€29	Low volume, single schema
Growth	€99	Batch endpoint, webhooks, multiple schemas
Scale	€299	Higher throughput, priority support

Indicative and not yet final — it is here so you can tell me whether it is plausible. Design partners from this page get launch pricing held.

What is actually true today

Built: nothing you can call yet. This is deliberate.
Being decided: where each provider physically runs inference, and their retention terms. Verified before anything ships, not after.
Not claimed: no certifications, no signed DPA, no uptime history. Those get built and published — they are not being implied here.

If a data-residency or compliance requirement has ever stopped you from using a document processing API, I would like to hear how that actually went — even if you would never buy this.

Get in touch

Leave an email and I will follow up personally. No sequence, no newsletter.