Paste a block of HTML, or choose an .html file from your computer, pick a page size, orientation and margin, and download a PDF. The markup is laid out by your own browser at the width of the printable area, photographed page by page, and the pictures are placed on PDF pages. Inline styles and <style> blocks apply, so a styled invoice, a report template or an email draft comes out looking the way the browser shows it.

The whole job runs in the page you are looking at. Your HTML is never sent to a server, and the PDF is built on your device. Because the output is an image of the rendered page, the text inside it cannot be selected, searched or copied, which is the trade this approach makes for working without a server.

Page size, orientation and margin

A4 and US Letter are offered in portrait or landscape. The margin, in millimetres, is applied on all four sides and the HTML is laid out to fit the width that remains, so a wide table at a 25 mm margin has less room than at 10 mm. Content that is taller than one page continues on the next.

Page breaks are placed on the nearest blank row of pixels above the natural cut, so a line of text is not sliced across two pages. There is no way to force a break at a particular element, and CSS page-break properties are ignored, because the page is rendered as one continuous strip and divided afterwards.

What is removed before rendering

The markup is passed through a sanitiser first. Script tags, inline event handlers, iframes, embeds and links to external stylesheets are stripped out. The cleaned document is then loaded into an isolated frame with scripting disabled, so even something the sanitiser missed has no way to run. JavaScript in your HTML will not execute, and anything it would have drawn on the page will not appear.

Styles written in a <style> block or a style attribute are kept. If your page depends on an external stylesheet, paste its rules into a <style> block in the document before converting.

Images and fonts from other sites

An image or web font hosted elsewhere loads only if that host allows cross-origin use. The browser refuses to let a page read pixels from an image whose server did not opt in, so such an image renders as a blank area in the PDF. Images embedded as data URLs, and any image or font served with permissive cross-origin headers, work as expected. The tool waits up to eight seconds for images and fonts to load before it takes the picture.

Quality and file size

Pages are rendered at twice the usual screen density, which reads well on screen and is acceptable in print for text at ordinary sizes. Each page is stored as a PNG when that is compact, which it is for text and flat colour, and as a JPEG when a page is dominated by photographs. A text-only A4 page is typically around half a megabyte, so a long document produces a far larger file than a text-based PDF of the same content would.

If you need a PDF whose text is selectable, use your browser's own Print to PDF command on the page instead. This tool exists for the cases where you have raw HTML and no page to open, or where you want a fixed page size and margin without touching print settings.

Frequently asked questions