* * * PDF TO MARKDOWN API * * *

PDF to Markdown API.

LLM pipelines want markdown, not PDF bytes. POST a PDF — born-digital or scanned — and get back clean markdown per page: headings, paragraphs in reading order, tables preserved. One call, $0.00075 per page.

START FREE — 50 PAGES
$0.75 / 1,000 PAGES
1,333 PAGES PER DOLLAR · NO CREDIT CARD
A WORKING STARTING POINT

Run a PDF. Inspect the Markdown.

Start with this one-page synthetic invoice. The returned Markdown below is from a live public-demo request on October 20, 2026. It retained invoice DEMO-1001, both line items and the USD 42.00 total. This is one reproducible example, not an accuracy benchmark.

# Python 3.10+. No dependencies. Download convert.py and sample-invoice.pdf above.
python3 convert.py sample-invoice.pdf --demo

# For your own documents and normal API limits:
export PENNYOCR_API_KEY=YOUR_KEY
python3 convert.py your-document.pdf --output ocr-output

The public demo reads the first PDF page only, has a 10 MB cap and a daily quota. The authenticated API accepts up to 50 MB and 500 pages per request. The script writes Markdown and the complete JSON response, including page results when returned. It does not retry requests automatically.

VIEW THE ACTUAL RETURNED MARKDOWN
# SYNTHETIC INVOICE

Test document only. No real customer or payment details.

**Sample Parts Company**

Invoice: DEMO-1001  
Date: October 20, 2026  
Bill to: Example Automation Team

| Item | Quantity | Unit price (USD) | Amount (USD) |
|---|---:|---:|---:|
| Widget | 2 | 12.00 | 24.00 |
| Gasket | 3 | 6.00 | 18.00 |

Subtotal: USD 42.00  
Tax: USD 0.00

**TOTAL: USD 42.00**

This sample tests a heading, invoice identifier, line items and a total.  
Compare extracted text to the original. OCR output is not accounting validation.

This request took 43.22 seconds end to end, including the upload and response. One measurement does not establish typical speed. Latency and extraction quality depend on the document and available engine. Check your own samples before choosing a production pipeline.

AT A GLANCE
UPDATED
PRICE
$0.75 per 1,000 pages ($0.00075 per page), prepaid
FREE TIER
First 50 pages free per account, no credit card
FORMATS
PDF, PNG, JPEG, WebP, TIFF
LIMITS
50 MB and 500 pages per request
OUTPUT
JSON with markdown (tables preserved) or plain text, per page and joined
INTERFACES
REST API, OpenAI-compatible chat completions, MCP server
CREDITS
Never expire; requests stop with a 402 when the balance runs out
01 / TRY IT

ONE ENDPOINT.

POST a file, get JSON back — the extracted text, per page and joined. PDF, PNG, JPEG, WebP or TIFF.

curl https://api.pennyocr.com/v1/ocr \
  -H "Authorization: Bearer $PENNYOCR_API_KEY" \
  -F "file=@report.pdf"

# $0.75 per 1,000 pages, first 50 free
02 / USE CASES

WHAT PEOPLE BUILD.

RAG INGESTION
Keep headings, paragraphs and tables together before splitting into chunks. Check retrieval quality with your own documents.
LLM CONTEXT
Feed agents readable documents: tables stay tables instead of becoming word soup.
SCANNED ARCHIVES
Image-only PDFs (no text layer) come out as real markdown, up to 500 pages per request.
PER-PAGE RESULTS
page_results[] keeps pagination, so citations can point at the right page.
03 / PRICE CHECK

HALF THE PRICE OF THE BIG CLOUDS.

Per 1,000 pages, public list prices, first tier.

PENNYOCR$0.75
AWS TEXTRACT$1.50
GOOGLE CLOUD VISION$1.50
AZURE DOC INTELLIGENCE$1.50
YOU KEEP50%
04 / FAIR QUESTIONS
GET YOUR API KEY →
QUESTIONS? HELLO@PENNYOCR.COM
HOW ARE TABLES REPRESENTED?
As markdown pipe tables, which parsers and LLMs read directly; line breaks inside a cell come back as <br>. Plain-text mode flattens them to aligned text if you prefer.
BORN-DIGITAL PDFS TOO?
Yes — every page is rendered and read visually, so mixed corpora (digital + scans) go through one pipeline with one output format.
WHAT DOES A 300-PAGE PDF COST?
$0.225. The first 50 pages are free.