Rotate PDF

Pick the angle that matches what you're seeing. If the page is rotated 90° clockwise from where it should be (top of the page is on the right edge), apply 270° to bring it back. If it's 90° counter-clockwise (top is on the left), apply 90°. If it's flipped completely upside down, apply 180°. Sideways scans from flatbed scanners are almost always 90° one way or the other depending on which way you fed the document — try 90° first, and if it goes the wrong direction, undo and try 270°. The result downloads instantly so trial-and-error is cheap.

This tool rotates every page by the same angle, which covers the common case of a fully sideways scan. For selective rotation — say, page 3 is upside down but the rest is fine — split the PDF first with [Split PDF](/tools/split-pdf), rotate the offending pages individually, then recombine with [Merge PDF](/tools/merge-pdf). It is a three-step workflow but still entirely browser-based and faster than installing Acrobat. If you find yourself doing per-page rotation often, that's a sign the original scan workflow needs fixing rather than every PDF.

No, neither. The rotation is metadata-only — pdf-lib updates each page's `/Rotate` entry in the page dictionary, which is just an integer (0, 90, 180, 270) telling PDF readers how to display the page. The actual content stream (text, images, vectors) is untouched, so there's zero re-rendering, zero re-compression, and zero quality loss. The file size shifts by a few bytes at most. This is fundamentally different from rotating an image (where pixels get resampled) or rotating in some older PDF tools that flatten and re-render the content. Your scanned text will stay just as crisp; embedded photos won't get re-encoded.

Yes. Modern printers and PDF readers (Adobe Reader, Preview, Chrome's built-in viewer, browsers, mobile readers) all honor the `/Rotate` flag and orient pages accordingly. The one edge case is some very old or unusual print drivers that ignore the flag — if you hit that, open the rotated PDF in any modern viewer and 'Print to PDF' to bake the rotation into the content stream, then send the new file to the legacy printer. For 99% of printing scenarios, the rotated file works directly.

Yes — the entire rotation runs in your browser using pdf-lib via JavaScript. Your PDF is never uploaded to a server. Open DevTools → Network tab before clicking rotate and you'll see zero outbound requests. This matters for contracts, medical records, tax forms, and anything else you wouldn't want sitting in a third-party log file. Compare to Smallpdf or iLovePDF, which require uploading the document to their servers (where retention policies vary and can change without notice) — JustUse.me processes everything locally and the file disappears from memory the moment you close the tab.