Organize PDF
Reorder, rotate and delete PDF pages with simple per-page buttons — the rebuilt file downloads straight from your browser. Your PDF never leaves your device.
Load a PDF — every page appears as a card with ↑ ↓ move, ⟳ rotate and ✕ delete buttons. All changes rebuild the download live.
No pages yet.
Organized PDF
Rotation is stored as the PDF's per-page rotate flag — a lossless metadata change; nothing is re-compressed. Deleting the last remaining page is refused: a PDF needs at least one page.
Your PDF never leaves your browser. Pages are reordered, rotated and copied into the new file in memory on your own device — no server call, no upload, no logging of your document. Disconnect from the internet after the page loads and it still works.
Reorder, rotate and delete PDF pages without uploading the file
Most "organize PDF" sites — iLovePDF, Smallpdf, PDF24 — do the work by uploading your file to their server and handing back the result. For a contract, a scanned ID or a medical record, that means a copy of a private document sits on someone else's machine. This tool does the same job with zero upload: PDF.js (the renderer built into Firefox) draws the page thumbnails, and pdf-lib rebuilds the output — both pinned, open-source libraries served from this page. The file is read locally with the browser's FileReader; nothing is transmitted, so there is nothing to leak.
Three operations, one button each
Load a PDF and every page becomes a card showing its thumbnail, its current position and the page number it had in the original file. Each card carries four buttons:
| Button | What it does |
|---|---|
| ↑ / ↓ | Move the page one position up or down |
| ⟳ | Rotate 90° clockwise — clicks accumulate: ×2 = 180°, ×4 = back to 0° |
| ✕ | Delete the page from the output |
The move buttons are deliberately not drag-only: single-step ↑/↓ works with a keyboard (Tab to the button, press Enter), with a screen reader, and on a phone where drag-and-drop between small tiles is fiddly. A worked example: a duplex scanner that flipped every even page upside down is fixed by clicking ⟳ twice on each even card — 180° each, stored losslessly — and a fax cover sheet at position 1 disappears with one ✕.
Rotation is a flag, not a re-render
The ⟳ button never touches the page's content stream. It sets the PDF's per-page /Rotate value — the same mechanism Acrobat's Rotate Pages uses — so viewers turn the page at display time. Text stays selectable, vector lines stay sharp, and scanned images keep their exact original bytes. That also means the output file is essentially the same size as the input (minus deleted pages): this tool compresses nothing. If the result needs to be smaller, run it through the PDF Compressor afterwards.
Every change rebuilds the download live
There is no Apply or Generate button to hunt for. Each move, rotate or delete replays the full list of changes against the untouched original and rebuilds the output with pdf-lib's copyPages, and the download button updates with the new page count and file size — typically well under a second for a normal document. Because rebuilds always start from the original, Reset changes returns to the exact loaded state at any time, and no amount of experimenting can corrupt the source file.
Honest limits
Encrypted PDFs are not supported: a file locked with an open password usually fails to load and is reported with a message, not worked around — unlock it in the application that created it first (files with only an owner-password permission lock usually still open). Deleting every page is refused, because a PDF must contain at least one page; you get an inline error instead of a broken download. Very large scanned files are limited by your device's memory — a multi-hundred-megabyte scan may thumbnail slowly on a phone. And this tool works on one file: to combine several PDFs use Merge & Split PDF, and to pull page ranges out into separate files use Split PDF.