Document to JSON API

Convert any document to JSON in one API call.

Upload a receipt, invoice, PDF, or image and get back clean, schema-stable JSON — vendor, date, totals, tax, and line items already parsed. No OCR pipeline, no manual mapping, no model to train.

Request
curl -X POST https://api.parseify.app/api/v1/documents \
  -H "X-Api-Key: bnj_live_xxxxxxxxxxxxxxxxxxxxxx" \
  -F "file=@invoice.pdf"
Response
{
  "vendor": "Byte Office Supplies",
  "invoiceNumber": "INV-2026-0419",
  "date": "2026-04-19",
  "totalAmount": 499.13,
  "vatAmount": 86.63,
  "currency": "EUR",
  "lineItems": [
    { "description": "A4 paper box", "quantity": 3, "unitPrice": 29.90, "totalPrice": 89.70 }
  ]
}

Any document in, the same JSON shape out

PDF to JSON

Multi-page PDF invoices and statements are read and converted to the same structured JSON schema as image uploads.

Image to JSON

Snap a photo of a receipt in JPEG or PNG format and get back structured JSON in seconds — ideal for mobile expense capture.

Receipt & Invoice to JSON

Vendor, invoice number, date, total, VAT, currency, and line items are extracted automatically into consistent, predictable fields.

Built for teams that need documents as data

Finance & Accounting Tools

Turn incoming invoices into JSON your ledger or AP workflow can consume directly, without manual re-keying.

Expense & Travel Platforms

Convert a photo of a receipt to JSON the moment it's uploaded, so expense line items populate automatically.

ERP & Procurement Systems

Match structured invoice JSON against purchase orders using reliable vendor, total, and tax fields.

Document to JSON, frequently asked questions

How do I convert a document to JSON?

Send a POST request with the file to Parseify's /api/v1/documents endpoint using your API key. The response comes back as structured JSON with vendor, date, amounts, and line items already parsed out — no OCR pipeline or manual mapping required.

Can I convert a PDF to JSON with an API?

Yes. Parseify accepts PDF, PNG, and JPEG files up to 20 MB and returns the same structured JSON schema regardless of input format, so your downstream code only has to handle one shape of data.

What JSON fields do I get back?

Every response includes vendor, invoiceNumber, date, totalAmount, vatAmount, currency, and a lineItems array with description, quantity, unitPrice, and totalPrice for each item on the document.

Do I need to train a model to convert documents to JSON?

No. Parseify's extraction model is already tuned for receipts and invoices, so there is no dataset to label or model to fine-tune — generate an API key and start sending documents immediately.