Feature

Invoice and Receipt Extraction

Pull header fields and full line items out of invoices and receipts.

What is Invoice and Receipt Extraction in Mahad OCR?

Available now. Invoices and receipts return both header-level fields and a line_items array, so totals can be reconciled against the individual rows. Input: Invoice or receipt image or PDF (multi-page up to 5 pages). Limitation: Very dense multi-page invoices beyond 5 pages are truncated.

Available now

What it does

Invoices and receipts return both header-level fields and a line_items array, so totals can be reconciled against the individual rows.

How it works

1Upload the invoice or receipt
2Header fields and line items are extracted together
3Amounts and dates are normalized
4Result is available as JSON for your accounting system

Sample result

Illustrative response shape using synthetic data.

{
  "type": "Invoice",
  "values": { "invoice_number": "INV-2026-014", "total": "1250.00", "currency": "USD" },
  "line_items": [
    { "description": "Consulting", "quantity": 10, "unit_price": 125.00, "amount": 1250.00 }
  ]
}

Supported inputs

Invoice or receipt image or PDF (multi-page up to 5 pages).

Fields returned

invoice_number, invoice_date, due_date
vendor / supplier and buyer details
subtotal, tax/VAT, total
currency
line_items[] — description, quantity, unit price, amount

API

POST /v1/documents/upload (optionally doc_type=invoice)

Security & privacy

Financial documents are tenant-isolated and audit-logged like every other upload.

Current limitations

Very dense multi-page invoices beyond 5 pages are truncated. Faded thermal receipts and handwritten totals reduce accuracy. No arithmetic guarantee — always reconcile totals in your system.