PDF to HTML converter

Convert a PDF to clean, semantic HTML: <h1>-<h3> headings, paragraphs, lists, links and code blocks that reflow on any screen, instead of hundreds of absolutely positioned text boxes. Copy the body into a CMS or download a full page. Nothing is uploaded.

PDF file

No PDF yet. You can also drop a file on this box.

Headings
Headers/footers

Pages (optional)
Output

Preview

What you get

A PDF has no paragraphs or headings, only pieces of text placed at x/y positions. This page reads them with pdf.js (Firefox's PDF engine) and rebuilds the document's structure, the same engine as our PDF to Markdown converter:

In the PDF	In the HTML
Larger text	`<h1>`, `<h2>`, `<h3>` from the biggest size down
Bold, italic, monospaced fonts	`<strong>`, `<em>`, `<pre><code>`
Link areas	`<a href>` on the words under them
•, –, 1., a) lines	`<ul>` and `<ol>`
Lines of a paragraph	One `<p>`, hyphenated line ends rejoined
Page headers, footers, page numbers	Removed (option)

No inline styles, no classes, no positioned <div>s: the result takes your site's CSS. "Full page" wraps it in a complete HTML document with a small readable stylesheet.

What it doesn't keep

This converter is built for getting the text back into an editable form, not for a pixel-perfect copy of the layout. Say so up front, so you can pick the right tool:

Content	Result
Tables	Cell text in reading order, not a HTML table For tables, use PDF to Excel or PDF to CSV.
Images, charts, logos	Dropped Save them with extract images from PDF.
Fonts, colors, exact positions	Replaced by plain HTML that your CSS styles
Two-column layouts	Column by column, in the order the PDF draws them
Scanned PDFs	No text: the pages are pictures Run OCR PDF first, then convert its output here.

Converters that keep the exact look (such as pdf2htmlEX) place every line in an absolutely positioned box with embedded fonts: it looks like the PDF but is hard to edit, isn't responsive and reads poorly on phones. Use those for a visual copy, and this page when the HTML is going into a website, a CMS, an email or an LLM.

How we checked it

We converted the "Attention Is All You Need" paper from arXiv, a Wikipedia article printed to PDF by Chrome, and the IRS Form W-9: the paper's title, Abstract and numbered sections came out as headings at the right level, and the Wikipedia article kept its 101 links.

Going the other way: HTML to PDF by API

If you generate documents yourself, it is usually easier to keep the source as HTML and make the PDF from it. The Shotmint API renders HTML or any URL to a PDF in Chromium, with your CSS, web fonts, headers and footers:

curl -X POST https://shotmint.creasework.com/v1/pdf -H "X-API-Key: YOUR_KEY" -H "Content-Type: application/json" \
  -d '{"html": "<h1>Invoice 1042</h1><p>Total: €120</p>"}' -o invoice.pdf

Each render is 1 credit. Free keys get 25 renders every month; €5 buys 2,000 more, and bought credits never expire. Try it without a key on HTML to PDF.

Get 25 free renders a month API docs

FAQ

Is my PDF uploaded? No. pdf.js reads it and the HTML is built in JavaScript on this page, so the file stays on your device. That matters for contracts, payslips and medical letters, which most online converters send to their servers.

Is there a page or size limit? No limit, no signup and no watermark. Very large PDFs depend on your device's memory; use a page range if your browser struggles.

Can it open password-protected PDFs? Yes, if you know the password. The page asks for it, and it stays in your browser.

Why are some headings at the wrong level? Levels come from font sizes, so a PDF that sets its title and sections at the same size can't be told apart. Set Headings to "None" and style them yourself.

PDF to HTML converter

What you get

What it doesn't keep

How we checked it

Going the other way: HTML to PDF by API

FAQ

Related tools