Send an image or PDF, get back structured JSON — vendor, line items, totals, tax, dates — in under 3 seconds. No ML pipeline to build, no per-model tuning, no six-figure enterprise contract.
One endpoint. Send a receipt or invoice (image/PDF), get structured, typed JSON back — ready to drop into your accounting software, expense tool, or bookkeeping SaaS. Built on modern vision-language models instead of a legacy OCR + rules-engine stack, so it handles messy phone-camera receipts and multi-page invoices without a training set.
POST /v1/extract
Content-Type: multipart/form-data
X-API-Key: {your_key}
file: receipt_costco_0912.jpg
doc_type: receipt
{
"doc_type": "receipt",
"vendor": "Costco Wholesale #117",
"date": "2026-09-08",
"currency": "USD",
"subtotal": 84.27,
"tax": 4.21,
"total": 88.48,
"payment_method": "VISA ****4471",
"line_items": [
{
"description": "KIRKLAND PAPER TOWEL",
"quantity": 1,
"unit_price": 21.99,
"amount": 21.99
},
{
"description": "ORG SPINACH 1LB",
"quantity": 2,
"unit_price": 4.49,
"amount": 8.98
}
],
"confidence": 0.97
}
Priced to be a no-brainer swap for indie devs and small SaaS teams currently paying enterprise OCR rates for volume they don't have.
Launching on RapidAPI. Join the waitlist for a launch-week discount code and to help shape the roadmap (batch upload, more doc types, webhook retries).