Convert invoice PDF to Markdown
Invoices are mostly a table of line items plus a few header fields. Convert them to clean Markdown, with the line-item table rebuilt and scanned or photographed receipts OCR'd, ready to parse or hand to an LLM.
A clean table you can actually use
Copying an invoice out of a PDF usually scrambles the line items into misaligned text, and a scanned receipt is just an image. Converting to Markdown fixes both: the line-item grid is rebuilt into a real, aligned Markdown table, and OCR turns a scan or a phone photo into selectable text. The header fields, vendor, invoice number, dates and totals, come through as text too. The result is not a rigid field-by-field export; it is clean Markdown that is trivial to parse, or to feed to an LLM that pulls out exactly the fields you need.
Convert an invoice in 4 steps
No account needed. Works for digital invoices and scanned or photographed receipts.
Open the converter
Install the Chrome extension or open the web app.
Add the invoice
Drag in the PDF, pick it from disk, or paste a URL. Scanned and photographed receipts work too.
Wait for the job
Status goes queued, processing, ready. Line items become a Markdown table and image text is OCR'd.
Copy, download or parse
Copy the Markdown, download a .md file, or pass it to an LLM to extract the vendor, totals and line items.
From a PDF invoice to parseable text
Aligned line items
Description, quantity, unit price and amount are rebuilt into a real Markdown table, not misaligned text.
Scanned receipts
A phone photo or a scanned receipt is OCR'd into editable Markdown.
Easy to parse
Clean Markdown is simple to read with a script, or to hand to an LLM for field extraction.
Many at once
Convert a folder of invoices with the batch workflow for bookkeeping or a pipeline.
Need specific fields as JSON? This tool gives you clean Markdown, which is the reliable first step. Pass that Markdown to an LLM with a short prompt, and pulling out the total, due date or tax becomes simple and consistent.
From invoice to fields, in two steps
First the PDF becomes structured Markdown; then a short prompt turns that into exactly the fields you need.
Step 1. The converter rebuilds the header and the line-item grid as clean Markdown:
# Invoice INV-2048 Acme Supplies Ltd · Issued 2026-05-12 · Due 2026-06-11 | Description | Qty | Unit price | Amount | | ------------------ | --- | ---------- | ------ | | Standis A4 paper | 12 | $4.50 | $54.00 | | Toner cartridge XL | 2 | $89.00 | $178.00| | Delivery | 1 | $15.00 | $15.00 | Subtotal $247.00 · Tax (10%) $24.70 · **Total $271.70**
Step 2. Because the data is already clean text, an LLM extracts the fields reliably from a one-line prompt:
Extract invoice_number, vendor, issue_date, due_date, total and the line_items (description, qty, unit_price, amount) as JSON from the invoice Markdown below.
Doing it in two steps is what makes the result dependable. Asking a model to read a raw PDF often drops or misreads cells; giving it an aligned Markdown table removes the layout guesswork, so the totals and quantities come out right. The same pattern works for receipts, purchase orders and statements.
Common questions
Can I convert an invoice or receipt PDF to Markdown?
Yes. The line items become a real Markdown table and the header text (vendor, dates, totals) comes through as text, including for scanned or photographed receipts via OCR.
Does it extract invoice fields as structured data?
It produces clean Markdown, not a fixed JSON of fields. That Markdown is easy to parse, or you can hand it to an LLM to pull out the vendor, total and line items reliably.
What about a scanned or photographed receipt?
OCR turns the image into selectable Markdown, so a phone photo or a scan of a receipt becomes editable text and a table rather than an image. See scanned PDF to Markdown.
Do the line items stay aligned?
Yes. Columns like description, quantity, unit price and amount are rebuilt into an aligned Markdown table rather than collapsing into a jumble of text.
Can I process many invoices at once?
Yes. Use the batch workflow over the API to convert a whole folder of invoices to Markdown for your bookkeeping or an LLM pipeline.
Is it free?
Yes. Convert anonymously in the browser on the free tier (3 slots, 10 MB files, a 15-minute time budget, 1-hour retention). Paid tiers raise every limit.