Tools & APIs · Document → Markdown
PDF to Markdown converter for AI, RAG and search.
Send a PDF, a scan, a Word file or a web page, and get clean Markdown or plain text back, ready for an AI workflow, a search index or a knowledge base.
$0.003per document converted
Failed documents are free. Scanned pages that need OCR are $0.01 each.
{
"urls": [
"https://arxiv.org/pdf/1706.03762v7",
"https://en.wikipedia.org/wiki/Markdown"
],
"outputFormat": "markdown",
"ocrMode": "auto",
"ocrLanguages": "eng+por"
}Paste this into the tool’s input on Apify, or send it to its API.
What you get
- Uses the PDF’s text layer first and runs OCR only on pages that need it. You can also force OCR on or off.
- Word, Office and HTML converted with headings, lists, links and tables kept.
- Web pages cleaned of navigation, cookie banners, footers and sidebars.
- Optional heading-aware chunks with overlap, token counts and page ranges for RAG.
- A status and error code per document, so failures are easy to handle.
How it works
Add your documents.
Links to PDFs, Office files, images or web pages.
Pick the output.
Markdown or plain text, the OCR mode, and the OCR languages, like English plus Portuguese.
Use the text.
One record per document, with page and word counts, ready for your pipeline.
Questions
Do scanned PDFs work?
Yes. Pages without a usable text layer go through Tesseract OCR. Pages with text are read directly and are free of OCR charges.
Does it keep tables?
Office and HTML tables come out as Markdown tables. Complex scanned tables depend on OCR quality.
What if a file fails?
Failed documents are not charged, and each one comes back with an error code you can act on.