Nothing is uploaded, and that is not a marketing line
Every tool in this section runs in your browser. The file you open is processed on your own device, it is never sent anywhere, and there is no server holding a copy. Close the tab and there is no trace.
That is worth more than it sounds. The usual free PDF tools upload your document to somebody else infrastructure, and people put contracts, invoices, medical letters and passport scans through them without thinking about it. Here there is nothing to think about, because there is nowhere for it to go.
PDFs
Merge, split, reorder, rotate and delete pages from one screen rather than five separate tools. Images and phone photos into a single tidy document. Fill in a form, draw a signature and flatten it.
Pages, and the four different things people mean by that
Pulling pages out, deleting pages, cutting one document into parts and putting a bundle back in order are four separate jobs that happen to share one engine underneath. Each has its own page here, because each is a different thing to need.
The order you type is the order you get, which turns extracting into reordering. Repeats work too, so asking for page one three times gives you three copies of it, which sounds odd until you need a signature sheet in a pack.
Compressing, which is two completely different jobs
A scan is a photograph of paper, often 300 dots per inch and twenty megabytes for ten pages. Redrawing those photographs smaller is enormously effective and loses nothing that was ever there. A document out of Word is already tiny and efficient, and doing the same thing to it makes the file BIGGER while permanently destroying the text layer, so it can never be searched or copied again.
Almost every free compressor rasterises both and mentions neither. This one looks at the document first, tells you which of the two you have, and marks the option that suits it. If you choose the destructive one on a document with real text in it, it puts the character count in front of you before it will do it.
What runs in a browser, and what honestly does not
Some file jobs genuinely cannot be done well on your own machine, and rather than shipping a bad version this site does not ship them at all. That list is shorter than it was, and it is worth saying which came off it, because the reason they came off is the whole argument for building this way.
Two that used to be on this list and are not any more:
- Putting a password on a PDF. This needed encryption WRITTEN into the file, and the usual browser libraries can read encryption without being able to produce it. A different engine solved it, so the tool now applies real AES-256 on your own machine, and the file never goes anywhere to have it done.
- PDF to Word. Pulling the text out was always easy. Rebuilding a faithful editable document out of absolutely positioned glyphs is the hard part, and it is why this sat unbuilt for a long time. It ships now with the honest caveat attached: it preserves the text and a workable layout, and it will not reproduce a complicated one exactly.
What is still genuinely off the list:
- Video compression. It caps out around a couple of hundred megabytes when the archetypal file is an eight hundred megabyte phone video, so it would fail on precisely the case you wanted it for.
- Anything that summarises or answers questions about a document. That means sending the file to a server and paying for every request, which cannot be given away without limits, accounts or a catch. It is the funnel the big PDF sites are built on and it is the one thing here that will not be copied.
Every one of those exists elsewhere as a free tool that uploads your file to somebody else’s server to do the work. That is the trade, and it is a reasonable one to make knowingly. It is not one this site will make on your behalf without telling you.