Merge several PDFs, pull out a page range, delete pages, rotate them, or split a file into one PDF per page — without uploading anything. The file is opened, edited and re-saved by your own browser; the download button at the end points at a blob on your disk image of the page, not at a URL we control. Close the tab and no copy exists anywhere.
Choose one or more PDFs to begin. Nothing is uploaded.
Merge copies whole pages into a new document — no re-encoding, so quality is bit-for-bit what you gave it and the output is roughly the sum of the inputs. Extract and delete use the page range box: 1-3,5,8- means pages 1 to 3, page 5, then 8 to the end; an open range like 4- always means “to the last page”, so it stays correct when the file changes. Rotate adds to the existing rotation instead of replacing it, which is why a page that was already sideways does not flip twice. Split writes one PDF per page into a ZIP, because browsers refuse a burst of automatic downloads; the ZIP is stored, not compressed, since PDF pages are already compressed.
Why there is no “compress PDF” button. Most of a PDF's size is images. Genuinely shrinking it means re-encoding those images — lossy, irreversible, and sometimes visibly worse. Rewriting the file structure typically saves only a few percent, and any site promising 90% either rasterises every page (turning your text into a picture) or is uploading your file somewhere. The honest options are: re-export from the source at lower image quality, or accept the size. Password-protected files cannot be opened here — without the password even a correct reader can only see ciphertext, so the tool says so instead of failing mysteriously. Digital signatures are invalidated by any edit, this one included; that is how signatures work, not a limitation of the tool.
Tukuk-OS