Combine files in order. Nothing is uploaded.
Drop in the files you want joined, drag them into the order you want, and download a single PDF. The merge runs in your browser, so large files are limited only by your own memory.
Drop the PDFs onto the panel or use the file picker. Nothing uploads — the files are read into memory in the tab and stay there. You can add more in a second pass without losing what is already listed.
Drag the rows. The merged document follows the list top to bottom, and the numbers on the left update as you rearrange, so what you see is the page order you will get. Remove anything you changed your mind about with the cross on the right.
Each row shows its page count. This is the cheapest way to catch the wrong version of a file before you merge, and it is worth a glance when several drafts of the same document are floating around.
The merge assembles a fresh document and copies the pages across whole. Text stays selectable, links keep working, and embedded fonts travel with their pages. Nothing is rasterised, so the output is as sharp as the inputs.
Inside the file, a PDF is a collection of numbered objects — fonts, images, page descriptions, metadata — with a table at the end recording where each one lives. A page object points at the objects it needs. Merging means copying the objects belonging to the pages you want into a new file and rebuilding that table so every reference still resolves.
Because the operation is a copy at the object level, nothing is re-rendered. The text is still text with its original font references. Vector artwork is still vector. An image embedded at 300dpi is the same image afterwards. This is why merging a hundred-page document takes a moment rather than a minute, and why the output is bit-for-bit as good as the input.
Some things live at the document level rather than the page level, and those need explicit handling. Form fields, digital signatures, bookmarks and outline trees, and document-level JavaScript belong to the original file. A signature in particular cannot survive, because it certifies the exact bytes of the document it was applied to, and a merged file is by definition different bytes. If the document is signed, merge first and sign afterwards.
A PDF keeps its pages in a structure called the page tree, a branching arrangement rather than a flat list, which lets a reader open page four hundred of a thousand-page document without walking through the first three hundred and ninety-nine. Merging rebuilds that tree from scratch with the pages in the order you set. This matters because a few other things hang off it: the document outline, or bookmarks panel, points at nodes in the original tree and does not survive the rebuild; page labels, which are what let a document number its front matter in roman numerals and its body in arabic, are also document-level and are reset. If the source documents had meaningful bookmarks, expect to recreate them.
Each source file carries its own embedded font subsets. Merge three documents that all use the same typeface and the output carries three copies of it, because the subsets are separate objects and may cover different character sets. This is why a merged file is occasionally larger than the sum of its parts. Running the result through the compress tool will collapse genuinely identical objects.
That file is almost certainly encrypted. Open it in any reader, print it back to PDF or save an unprotected copy, and merge that. Some files are also structurally damaged — produced by a broken exporter — and a save-as from a reader will usually repair them.
The list order is the output order. If a file itself has pages in an odd sequence, the merge preserves that. Use the split tool to reorder within a file first — listing pages as 3, 1, 2 outputs them in exactly that order.
Duplicate font subsets and repeated images across the source files. Run the output through PDF Compress, which rewrites the structure and drops redundant objects.
Interactive fields are document-level objects and do not survive a page-level copy. If you need a fillable form combined with other documents, fill and flatten the form first, then merge the flattened version.
Everything happens in memory in one tab. A very large set on a machine with limited RAM will struggle. Merge in batches — combine four files, then combine the results — which keeps peak memory lower.
Acrobat does everything this does and a great deal more: redaction, OCR, preflight, proper form editing, signature workflows. It also costs a subscription. If merging is the only thing you need, paying monthly for it is poor value. If you handle documents professionally, Acrobat earns its price.
The large free PDF sites are convenient and generally competent. The trade is that your document goes to their server, sits in temporary storage for some period described in a privacy policy you did not read, and most of them cap file size or daily operations to push you toward a subscription. For a restaurant menu this is fine. For a contract, a medical record, or anything under NDA, it is a genuine disclosure. Doing the work in the browser removes the question entirely.
macOS Preview merges PDFs by dragging pages between sidebars, and it is excellent — free, local, already installed. If you are on a Mac, use it. Windows has no built-in equivalent, which is most of the reason tools like this one exist.
There is no server limit because nothing is uploaded. Very large sets may slow down on older machines, since the whole document is assembled in memory.
Yes. Files can be dragged into any order before merging, and individual pages can be excluded.
Encrypted files must be unlocked first. You will be prompted for the password before the file can be added.