PDF viewer

Drop the file in and read it. There's nothing to install, no account, and the document doesn't leave your machine.

Two ways to move through it. One page at a time, with arrows and the keyboard, which suits a document you're reading properly. Or the whole thing as a scroll, which suits skimming and lets the browser's own find work across what's on screen.

Pages can be reordered and saved, which is the one thing here that changes the file rather than just showing it.

Why it stays on your machine

Every page is rendered in the browser. Nothing is uploaded, and there's no server holding a copy afterwards.

That's the practical difference from most online PDF tools, and it matters because of what people open: contracts, payslips, medical letters, bank statements, tax documents. Handing those to a service to save a download is a poor trade, and it's a trade most viewers make quietly.

The one exception on this site is the Office viewers, which genuinely can't work in a browser and say so on their own pages.

What a viewer can't tell you

Whether the text is real text. A scan looks identical to a typeset document until you try to select something. If your cursor draws a box rather than highlighting words, it's pictures of pages, and getting the words out needs recognition rather than extraction.

Whether it's been altered. A PDF carries no visible history. Edits made before you received it are invisible, and a document that looks untouched may not be.

Whether a signature means anything. A drawn or pasted signature is a picture. A cryptographic signature is a different thing entirely, and reading it requires software that checks the certificate rather than displays the page.

Reordering pages

Moving pages here rewrites the document properly rather than rearranging pictures of it. The real pages are copied into the new order, so the text stays text and stays searchable.

Pages are labelled by position rather than by the number they came from, because once something has moved, the original number is misleading.

The saved file is a copy. The original you opened is untouched, which is the behaviour you want when a reorder turns out wrong.

Files that won't open

Password protected documents. A viewer can't open what it can't decrypt. If you know the password, opening it in a reader and removing it first is the route.

Damaged files. A truncated download or an interrupted transfer produces a file that starts rendering and stops. If a document opens partially, that's usually why, and re-downloading it is the fix.

Very large documents. Everything is held in memory, so a several hundred megabyte file can struggle where a smaller one is instant. Splitting it is the practical answer.

Reading long documents

The scroll view is better for finding something, because the browser's find works on the pages that have rendered.

The page-at-a-time view is better for reading, and the arrow keys move through it without reaching for the mouse.

For a document you'll come back to repeatedly, downloading it and using a desktop reader is still the better experience. This is for opening something once, on a machine that may not have a reader installed, without giving the file to anyone.