Convert Markdown to PDF

Create print-ready PDFs from Markdown in your browser — selectable text, working links, real tables, free and private.

Free forever · Runs in your browser · No signup · Fidelity report included

Last verified: 2026-10-25 · engine v0.5.0 · changelog

How to convert Markdown to PDF (3 steps)

1. Add inputPaste Markdown or choose a .md file
2. Review previewReview the live preview and the conversion fidelity report

Full workflow: getting AI chat output into documents

3. Download PDFDownload your PDF — text stays selectable, links stay clickable

Example: Markdown → PDF

The same kitchen-sink document as the Markdown → Word tool — so you can compare exactly where the two engines agree and differ.

Input (Markdown)
# Quarter in review

Revenue grew **18%** — see [the report](https://example.com/report).

| Region | Q4 |
|---|---:|
| EMEA | 1.4M |

1. first step
2. second step

- [x] audit
- [ ] sign-off

> quoted wisdom

```bash
deploy --region emea
```

A footnote.[^1]

$$E = mc^2$$

![chart](https://example.com/q4.png)

<p>raw html block</p>

[^1]: The source note.
Output (pdf)

Content preview of the exported PDF file — the .pdf carries this structure; the fidelity note below states the exact per-element treatment.

pdfcontent preview

Quarter in review

Revenue grew 18% — see the report.

RegionQ4
EMEA1.4M
  1. first step
  2. second step
  • audit

  • sign-off

quoted wisdom

deploy --region emea

A footnote.1

E=mc2

[image: chart]


  1. The source note. ↩
Page 1 of 1

Fidelity report for this exact input: Measured on this exact input: 10 native, 2 approximated (math typeset with real sub/superscripts and symbols; footnotes as an end-of-document Notes section), 2 dropped (remote image — never fetched; raw HTML). Fidelity 70/100, shown before you download.

What converts natively (and what doesn't)

FeatureResult in the PDF
Headings 1–6Native — Sized and bolded on the page (PDF outline/bookmarks not generated yet)
GFM tablesNative — Real table with a shaded header row that repeats across page breaks
LinksNative — Clickable, coloured in the PDF
Ordered and bulleted listsNative — True numbered and bulleted lists, nested
Code blocksNative — Monospace with a shaded background and preserved spacing
Task listsNative — Printed checklist glyphs ☑/☐ — the correct form for a static PDF
BlockquotesNative — Indented, italic block styling — the typeset form of a print blockquote
Math (LaTeX)Partial — Typeset with real sub/superscripts, √ and ∑/∫ glyphs (engine v0.4.0); stacked fractions and OMML remain Word-only
ImagesPartial — data: PNG/JPEG images are embedded; remote/relative files are never fetched (privacy) and stay flagged in the report
FootnotesPartial — Collected into a Notes section at the end of the PDF (not paged footnotes)
Raw HTMLNot supported — Removed for safety
Front-matter (YAML/TOML)Native — title/author/subject/keywords map to the PDF document info; other keys are listed in the report
Mermaid code blocksPartial — Rendered as code blocks; diagram rendering is planned

Full element-by-element ground truth, with test dates: format support.

What this tool does not do

  • No: Remote/relative images are never fetched (privacy); data: PNG/JPEG images are embedded
  • No: PDF bookmarks / document outline
  • No: Automatic table of contents
  • No: Letter/landscape page sizes — the engine renders A4 portrait
  • No: Fonts beyond the embedded Roboto subset (Latin, Greek, Cyrillic)

What people use Markdown to PDF for

Print-ready reports and handoutsTurn specs, meeting notes and documentation into A4 PDFs with repeating table headers and page numbers — ready to print or attach, with nothing uploaded in the process.
Share README files outside GitHubConvert a project README or changelog into a PDF that opens anywhere — links stay clickable for readers who do not live in a code editor.
Archive AI conversations as documentsSave a ChatGPT or Claude session with its tables and code blocks as a clean, dated PDF instead of a folder of screenshots.

Frequently asked questions

Is the PDF text selectable?

Yes. The PDF contains real embedded text and embedded fonts — you can select, search, and copy from it.

Do tables repeat their header on new pages?

Yes. Long tables that cross a page boundary repeat their header row, like a properly typeset document.

Is my content uploaded anywhere?

No. Everything runs locally in your browser — the Markdown you paste or choose never leaves your device, and no conversion request is sent to any server. You can watch the network tab stay empty while converting.

What page sizes are supported?

The engine renders A4 portrait — the universal print default — with page numbers in the footer. Letter and landscape presets are planned.

Are images supported?

As of engine v0.3.0, data: PNG/JPEG images are embedded directly into the PDF. Remote images are never fetched for privacy — those references are flagged in the fidelity report instead.

Which fonts does the PDF use, and are they embedded?

The PDF embeds a subset of Roboto covering Latin, Greek and Cyrillic, so text renders identically on machines without the font installed and stays selectable and searchable.

Can I control page size or margins?

The engine renders A4 portrait with comfortable margins and page numbers in the footer. Letter, landscape and margin presets are on the engine roadmap.

Where did my footnote go in the PDF?

Since engine v0.3.0, footnotes are collected into a “Notes” section at the end of the PDF, marked with the same superscript numbers. A true per-page footnote layout is still ahead — the fidelity report tells you this before download.

Does YAML front-matter become PDF metadata?

Yes — title, author, subject and keywords from YAML front-matter land in the PDF document info (visible in a PDF reader's properties), and the fidelity report lists any front-matter keys without a metadata equivalent.

Is there a file-size or page limit?

No artificial limits on either. The engine renders A4 portrait in your browser, so long documents simply produce more pages — the practical ceiling is device memory, not a server quota.

Can I make a resume or CV from Markdown?

Yes — many people do. The PDF has selectable, searchable text with embedded fonts, which is what applicant tracking systems parse. Note the engine renders A4 portrait with page numbers in the footer.

Will the PDF look the same on every machine?

Yes — the PDF embeds a Roboto subset covering Latin, Greek and Cyrillic, so text renders identically everywhere and stays selectable. No font substitution happens on the reader's machine.

Related tools

Related searches: markdown to word · markdown editor · word to markdown · markdown to image