Skip to main content

PDF to Markdown

Get a PDF's text as Markdown for notes, editing or an AI prompt.

Drop your PDF here

or click to browse

PDF files up to 50 MB

Converting…

Your Markdown

· ·



                
            

Layout is not rebuilt: tables come out as plain lines and large text does not become a heading. AI OCR sends page images to an outside provider and has a page limit.

  • Markdown with page headings
  • Copy or download a .md file
  • Optional AI OCR for scans

How do I turn a PDF into Markdown for notes or an AI tool?

Upload a PDF and get Markdown with a heading for each page, ready to copy or download as a .md file. You get the text only: bold, links and images are left out, and tables come out as plain lines. When AI OCR is available, scans are read by it, up to the first 8 pages by default.

How do I convert a PDF to Markdown?

Upload the file and the Markdown appears in your browser. Copy it, or download it as a .md file named after your PDF. The page also shows the page count, the word count and whether the text was read directly or by OCR.

The result is plain, clean text that pastes well into a notes app, a wiki or an AI assistant. It is not a copy of the page design.

What does the Markdown look like?

The file name becomes the top heading, and each page gets its own heading so you can find your place against the original. A page with no text is marked as such, which keeps the page numbers in step. This is page 1 of a two-page report we converted:

Markdown output, page 1

# pump-station-inspection ## Page 1 Harbour Works Ltd - internal Pump Station Inspection Left column. The intake screen was cleared on arrival. Flow resumed within ten minutes and the level alarm reset without a fault code. Right column. Bearing temperature on pump two read high. The unit was isolated and a replacement bearing was ordered. Asset Reading Status Pump 1 41 C OK Pump 2 78 C Isolated Screen Clear OK

Do headings, tables and formatting come through?

No. The converter reads the words, not how they look. In the sample above, "Pump Station Inspection" was a large printed title, and it came out as an ordinary line.

  • Headings: large or bold text is not turned into Markdown headings. The only headings are the file name and the page numbers.
  • Tables: each row comes out as one line with tabs between the cells. Most Markdown apps show that as running text, so add | separators by hand if you need a real table.
  • Bold, italics, links and lists are not carried over.
  • Images are left out. Use the image extractor if you need them.
  • Line breaks fall where they fell on the page, and columns come out one after the other.

Can it convert a scanned PDF?

Yes, when AI OCR is available. If the file has almost no real text, it is treated as a scan and an AI model reads the pages from images. By default only the first 8 pages are read, and the result tells you how many were processed. For a longer scan, split the PDF into parts of 8 pages or fewer and convert each.

OCR can misread names, figures and dates. Check anything you will rely on against the original page.

The choice is made for the whole document, not page by page. One scanned page inside a typed report is not sent to OCR and shows up as a page with no text.

Is my file sent to an AI provider?

Only when it looks like a scan. A PDF with normal text is converted entirely on our server and nothing leaves it. For scans, images of the pages being read go to the OpenAI API. A PDF that needs a password, or has permission restrictions set by its owner, is refused before anything is read, so it is never sent.

The Markdown is sent straight to your browser. The uploaded PDF and any page images made for OCR are deleted when the request finishes.

Why did the conversion fail or come back empty?

  • "OCR could not read this PDF right now." The file had no usable text and OCR could not run or failed.
  • A message that the PDF is password-protected, damaged, or has permission restrictions. The file is refused before anything is read, and the refusal does not use your daily allowance. Unlock it with its password, using the unlock tool, and try again.
  • "No readable text found in this PDF." Neither the text in the file nor OCR produced anything.
  • Pages marked as having no text in an otherwise good result are usually scanned pages inside a typed document. Split those pages out and convert them separately so OCR reads them.

How many PDFs can I convert a day?

Guests can try one file. Verified Free accounts can convert up to 10 PDFs a day within their shared monthly upload allowance. Pro and AI accounts can convert 100. Each file also uses an upload from the shared allowance. When it runs out you see "You have used today’s conversion allowance."

Last updated . Written from the converter's source code and from test PDFs we built for the purpose.

Do more with a free account

Keep your report history, share reports by link, check several PDFs at once and use the API.