Free PDF tools that never upload your files
Merge, split, delete, extract, reorder, rotate, compress, number, resize, strip metadata, extract text, convert to images, convert images to PDF — fourteen tools, all running in your own browser. No upload queue, no waiting for a download, and nothing to hand over.
All tools
Why there is no upload button
Most online PDF tools work the same way: your file goes up to a server, gets processed there, and comes back down. That design has two costs. The file spends time on someone else's disk — and contracts, ID scans, medical letters and payslips are exactly the documents you would rather not put through that. The second cost is time: a 60 MB scan has to travel twice before you see anything.
Browsers can already do this work. PDF.js parses and renders pages; pdf-lib rewrites the document structure and saves it back out. Merging, splitting, rotating or numbering pages needs no server at all — it needs a few hundred milliseconds of your CPU. That is the whole architecture here: the page downloads a JavaScript library, and the file never leaves the tab.
The practical consequences are worth spelling out. Speed depends on your machine, not your connection. The size ceiling comes from available browser memory — roughly 100 MB per file on a desktop, 30 MB on a phone — not from a subscription tier. And the site keeps working when the network drops, because after the first load there is nothing left to fetch.
What these tools do well
Page-level operations are lossless. Merging copies page objects into a new document tree; splitting, deleting, extracting and reordering do the same in reverse. Rotating writes a rotation flag instead of re-rendering. Nothing is decoded and re-encoded, so a merged document is as sharp as its sources and roughly the size of their sum.
Text extraction reads the text layer the PDF already contains. It returns the characters in reading order; it does not rebuild columns, tables or indentation. If you need the layout, you need the original document, not a PDF.
Where compression has a trade-off
Compression is the one operation that cannot be free. Lossless optimization rewrites the file structure and removes redundant objects: the quality is untouched, but the saving is modest, and files exported by modern tools often shrink by nothing at all — in that case you get the original back unchanged. Scan compression renders each page to an image and rebuilds the PDF around it: files typically drop to a third of their size, but the text stops being selectable or searchable. Each mode says this on the page before you run it.
What is deliberately missing
There is no PDF-to-Word converter and no OCR. Both need a model that reconstructs layout — heading levels, table cells, image placement — and a browser-only implementation produces the kind of document that takes longer to fix than to retype. There is also no password removal: opening a file you do not have the password for is not a feature worth building, and the tools that offer it attract the wrong traffic entirely.
Longer reads
Three guides take the topics a tool page has no room for: where your file actually goes and where local processing stops being enough, what really makes a PDF large and which reduction method fits which document, and what a PDF carries besides its pages.