Merge PDF files
Add PDFs, drag them into the order you want, and download one combined document.
How it works
Merging is a structural operation, not a conversion. A page in a PDF is already a self-contained object (a content stream plus references to the fonts, images and colour data it draws with), so combining documents means copying those objects into a new file and renumbering them, then writing a fresh cross-reference table. Nothing is rasterised, re-encoded or re-laid out. Text stays selectable and searchable, vector line art stays vector, embedded fonts travel with the pages that use them, and a 300 DPI scan is the same 300 DPI scan on the other side.
Page geometry is copied too, so mixing an A4 report with Letter pages and a rotated scan keeps each page at its own size and orientation rather than resizing everything to match, and a print-ready file whose visible page is inset from the sheet keeps that inset. It is also why merging is quick even on large documents: the work is bookkeeping, not rendering.
Order lives in the file list
Order is the entire decision here, and it belongs in the file list rather than in a settings field. Drag the rows to arrange them; the merge follows that order exactly, top to bottom, with every page of the first document placed before the first page of the second. This matters because selecting files rarely gives you the order you meant: most pickers hand over alphabetical or click order, which puts chapter-10.pdf ahead of chapter-2.pdf. Adding more files appends them to the end of the list instead of replacing it, so a long sequence can be assembled in several passes.
A file that appears twice in the list is included twice, in both positions: a repeated row is read as a deliberate instruction, which is what you want when the same cover sheet or terms page belongs in two places, and it is visible in the list if it was not. Repeating a file costs little beyond its pages, because a byte-identical input is read and stored once however many times it appears. The same 556 KB document included twice came out at 335 KB. Rearranging individual pages is a separate job, and organize PDF pages does that.
What does not travel with pages
Not everything in a PDF travels with its pages, and it is worth knowing what does not. The merged file gets fresh document properties: title, author, subject and keywords are not inherited from any source, because with twenty inputs there is no defensible way to pick one. Each source's own bookmark tree is dropped as well; with the bookmark option left on, the merged document instead gets one top-level entry per source, named after its filename and pointing at that document's first page.
Interactive form fields are the sharp edge. Field widgets are copied and still look correct, but the combined document does not re-register them as a form, so filled values survive visually while the fields stop being editable. Even if they stayed editable, two documents that each define a field called Name would collide on one value. Flatten or finish forms before merging. Encryption does not carry over either: a source restricted against printing or editing contributes its pages to a merged file that carries no such restriction. Web links keep working; links that jump to another page within a source document generally do not survive the renumbering.
Why a merge can grow
Expect roughly the sum of the inputs, and be ready for more, because a merged PDF can come out larger than the files that went into it. Measured on real documents rather than test fixtures: two book-length PDFs of 441 KB and 274 KB merged to 1.3 MB, 92% above the arithmetic total, while a 50 KB resume and a 31 KB invoice merged to 65 KB, 20% below it, and a 2.6 MB scan combined with that resume finished 1% smaller.
The difference is structural rather than a matter of content. A PDF may hold its objects and its cross-reference table in a compact compressed form, and rebuilding the document writes that structure back out plainly: the inflating files here are compressed thoroughly enough that even their own producer names sit inside those streams. Files written the plain way, including the scans and the exports from Microsoft Print to PDF, Google Docs and iText measured above, rebuild neutrally or shrink, because unused objects and each source's trailer are left behind.
The extra bytes are recoverable
None of that is lost detail: the pair that grew 92% was only 3% larger than its inputs once both were gzipped, which is what shows the extra bytes are recoverable rather than spent, and why compress PDF is a real remedy here rather than a consolation. Two different files never share that way, since nothing compares the insides of one document against another, so a font or logo present in both is kept twice.
Questions
Are my files uploaded?
No. Each PDF is read into memory by the browser, combined by a PDF library running in a Web Worker in this tab, and written back out as a download. Open the Network tab in developer tools and merge something. There is no request carrying your documents, because there is no server to carry them to.
How many files can I merge, and how large can they be?
Each file is capped at 200 MB and rejected at selection time if it is over, with the size and the limit stated. The practical ceiling on the total is your browser tab's memory rather than a fixed number: documents are loaded and released one at a time so that only one source is resident at once, which is what makes twenty large PDFs workable on a desktop. A phone will manage considerably less.
Can I merge a password-protected PDF?
It depends which kind of protection. Many PDFs (tax forms, statements, ebooks) are encrypted only to restrict printing or editing and open in any viewer without a prompt; those merge normally, because an empty password is all they require. A file that genuinely asks you for a password before it will open cannot be read, and this tool never asks you for one. In that case the merge stops and names the file, so you know which one to unlock and re-save first.
Why did the form fields stop working after merging?
A PDF form is not stored on the page; the page holds the visible widgets while the document catalog holds the field definitions that make them interactive. Copying pages brings the widgets but not that catalog entry, so the fields render with whatever values they had and are no longer editable. If you need a fillable form to remain fillable, keep it as its own file, or fill and flatten it before merging.