Merge PDFs Without Uploading Them

Order is the job

Files merge top to bottom in list order, and the reorder arrows exist because file pickers do not respect meaning: the operating system usually returns files alphabetically, which puts page-10 before page-2 and an appendix before the contract it belongs to. Check the list before merging — it is the single most common complaint about merge tools, and the evidence that something went wrong arrives much later, page ten deep in a document.

The output name follows the first file rather than a generic merged.pdf: contract-signed.pdf plus two annexes becomes contract-merged.pdf, which still says which contract it is six months from now.

Why there is no page limit and no watermark

Page caps, watermarks and daily limits exist on other sites because every merged page costs their server bandwidth and storage. Here the cost is paid by your own CPU, in your own tab, so there is nothing to ration and no paid tier to push you toward. The limits that remain are physical rather than commercial:

• 100 MB per file and 250 MB per selection — a PDF is held in memory several times over during a merge, and past that a browser tab dies mid-merge with the work lost. The tool declines instead of letting that happen.
• 100 files per merge — enough for a year of monthly statements; beyond that the list itself stops being usable.
• Encrypted PDFs are declined — a password-protected document cannot be read without its password. Unlock it first if you have it; if you do not, nothing can open it.

Browser versus desktop: an honest comparison

Neither column wins everywhere. For the everyday merge — a handful of documents under a few hundred pages — the browser wins on privacy and friction. For a multi-gigabyte scan archive or a nightly automated job, a desktop tool is the right instrument, and pretending otherwise would be the dishonest version of this page.

What merges cleanly — and what deliberately does not

Every page is copied with its complete dependency graph — fonts, images, colour spaces, vector content — and every object reference is renumbered into the destination document. Page content, page dimensions, and the annotations and form fields that belong to a page all survive exactly; they travel with the page.

Bookmark outlines are a different case, and the honest answer is that they do not merge. An outline is document-level structure with a single root: two documents each have their own tree, and stitching them produces destinations pointing at the wrong pages. Rather than emit a broken outline that looks right until you click it, the merge emits none — the same policy as the document-level AcroForm dictionary. If outline preservation is the actual requirement, that is the desktop column above.

The file-size surprise

Merge a 2 MB and a 3 MB PDF and the output is often not 5 MB — usually a little less, occasionally noticeably less. A PDF carries more than its pages: unreferenced objects left behind by earlier edits, old revisions from incremental saves, duplicate copies of the same embedded font. A merge writes the destination document fresh, so shared resources are written once and dead weight is not carried across. The flip side: a merge is not a compressor — page content is copied, not re-encoded — so a file that was already lean merges to roughly the sum of its parts.