Skip to main content

PDF to Markdown Converter

To convert a PDF to Markdown, upload the PDF and choose Markdown (MD). Our server extracts the text, turns larger font sizes into #, ## and ### headings, recognises bulleted and numbered lists and returns a GitHub-flavoured .md file. Tables in the PDF are not rebuilt as Markdown tables; their text arrives as paragraphs. Scanned PDFs are not supported because there is no OCR.

  • Free, no sign-up
  • Files deleted within 30 minutes
  • Up to 100 MB

Files are permanently deleted from our servers within 30 minutes.

How to convert PDF to Markdown

  1. Upload your PDF file

    Drag your PDF file onto the upload area or browse for it (up to 100 MB).

  2. Keep Markdown as the target

    Markdown is already selected in the menu; you can pick another format if you need one.

  3. Convert and download

    Press “Convert” and the download of the converted file starts automatically.

What is preserved?

Kept

  • Body text and paragraphs
  • Headings detected from font size (#, ##, ###)
  • Bulleted and numbered lists

Changed

  • Lines starting with #, -, > or 1. are escaped so they stay literal

Not carried over

  • Table structure: cells become paragraphs
  • Fonts, colours and page layout
  • Full-page background images

When to use it

Preparing documents for LLMs and RAG pipelines

Before you give PDFs to ChatGPT, Claude or a retrieval-augmented generation (RAG) system, converting them to Markdown strips page-layout noise while keeping the heading hierarchy, which makes chunking and citing sections much easier.

Moving manuals into docs-as-code and note apps

When migrating PDF manuals to GitHub, a GitLab wiki, Obsidian or Notion, you can start from a .md file whose headings and lists are already marked up instead of formatting everything by hand.

Limitations to know

  • No OCR for scanned PDFs.
  • Heading detection relies on font size; bold body-size headings may stay paragraphs.
  • Encrypted PDFs are rejected.

About the formats

What is PDF?

PDF (Portable Document Format) is a fixed-layout document format created by Adobe and standardised as ISO 32000. Every piece of text, image and line is stored with its position on the page, so a PDF looks the same on every device and prints reliably. The trade-off is editing: a PDF stores positions on a page, not paragraphs and tables. Scanned PDFs contain only page images and need OCR before any text can be extracted.

What is Markdown (MD)?

Markdown is a lightweight markup language designed by John Gruber in 2004. Formatting is written as simple symbols inside plain text: # for headings, ** for bold, - for list items and | for tables. Markdown (.md) files are everywhere in GitHub READMEs, technical documentation, note-taking apps and AI assistant output. The most widely used flavour is GitHub Flavored Markdown (GFM), which adds tables, task lists and strikethrough.

Frequently asked questions

How are headings detected in the PDF?

The converter compares the font size of each text block with the body text. Blocks that are clearly larger become #, ## or ### headings depending on how much larger they are. Headings set in bold at body-text size may stay as normal paragraphs, which you can fix in the .md file.

Do PDF tables become Markdown tables?

No. A PDF stores only positioned text, not rows and cells, so the converter writes table contents as paragraphs. If you need real Markdown tables and you also have the document as Word or DOC, convert that file to Markdown instead, because tables in Word files are preserved.

Is the output good for ChatGPT and other LLMs?

Yes, that is a common use. Markdown is compact and easy for language models to parse: headings mark the sections and list markers mark the items. Upload the .md file or paste its contents. Just remember that scanned PDFs produce no text, since no OCR is performed.

Is converting PDF to Markdown free?

Yes, completely free. There is no account, subscription or watermark — just upload your file and convert it.

Is my file safe and when is it deleted?

Your file is transferred over HTTPS and stored on our server under a random code. The upload and the output are permanently deleted within 30 minutes, and the download link only works for whoever holds your unique code.