Merge PDF files — without uploading them

Combining PDFs is the single most common reason people hand a document to a website they know nothing about. The files involved are rarely trivial: signed contracts, invoices, medical records, passport scans assembled for a visa application, a mortgage pack. Every mainstream free merger works the same way — your files are uploaded, processed on their infrastructure, and held for some retention window you have to take on faith.

This merger never uploads anything. The combining is done by pdf-lib running in a Web Worker inside your own tab, and the site is static files with no backend that could receive a document. A Content-Security-Policy blocks third-party requests on top of that. The practical test takes ten seconds: load this page, disconnect from the internet, and merge anyway. It works, because your machine was doing the work the whole time.

Nothing is retained either, because nothing was ever stored — close the tab and the files are gone from memory. There is no "your files are deleted after an hour" promise to believe, because there is no upload in the first place.

Drop two or more PDFs, or pick them with the file chooser. Each one appears as a row with a thumbnail of its first page, so you can tell similarly-named exports apart without opening them. Files are combined strictly top to bottom: the first row becomes page one of the result.

Reorder by dragging rows, or with the arrow buttons if you would rather not drag — that matters when you are assembling something order-sensitive like a contract followed by its appendices. Nothing is written until you click Merge, so rearranging costs you nothing and you can back out at any point.

Need to reorder pages inside a document rather than whole files, or drop a few pages first? Rotate and reorder PDF pages gives you a page-level grid, and Split PDF extracts the ranges you want. Both run on your device too.

Free online mergers cap uploads — commonly somewhere between 10 MB and 100 MB — because every megabyte costs them bandwidth and CPU. That ceiling is a business constraint, not a technical one, and it is exactly what pushes people toward a paid tier at the worst moment.

There is no fixed limit here. The only real bound is your browser's available memory, since the documents are held in your own RAM while being combined. In practice that is far more headroom than the hosted services allow; we show a heads-up once the combined input passes roughly 300 MB, purely so a very large job does not surprise you by getting slow. A modern laptop merges a few hundred megabytes without difficulty.

Password-protected PDFs cannot be merged directly. They are detected and flagged in the list rather than failing silently partway through, and the other files in your list are unaffected — remove the password first with Unlock PDF, then merge. If you want the combined result protected afterwards, Protect PDF will encrypt it, again without an upload.

Merging preserves the actual page content rather than flattening pages into images: text stays selectable and searchable, and vector graphics stay sharp at any zoom. Bookmarks and form fields from the source documents are not carried across, which is the usual trade-off for a fast client-side merge.

Want the result smaller for email? Compress PDF downsamples the images inside it while keeping text selectable, and also runs entirely on your device.