Useful for scans, contract attachments, handouts and multi-part reports: it joins several independent PDFs into one file in the order you choose. Source PDFs are read and combined entirely in your browser — nothing is uploaded, which suits documents that must not leave your machine.

Merging is concatenation, not re-layout: every page is copied over from its source file untouched, so text, images and vector content are never recompressed and never lose sharpness. The output has exactly as many pages as all sources combined.

1. 01Choose filesclick to select or drag in multiple PDFs; the list shows them in the order they were added.
2. 02Arrange the orderdrag list items around — the list order is the page order of the final file.
3. 03Run the mergeclick merge and let local processing finish; larger files and page counts take longer.
4. 04Preview and downloadcheck the page order in the preview, then download the exported PDF.

For example, combine cover.pdf, body.pdf and appendix.pdf in that order into one delivery file:

If the order is wrong you do not have to re-select anything — just drag the list items. The comparison below shows the list before and after dragging the appendix from first to last position:

Internally a PDF is a page tree; merging copies each source file's page objects and their resources into a new tree and numbers them in list order¹ — the full object model, page tree and the reason bookmarks get lost are explained in PDF file structure basics (an on-site guide, not yet published). What matters for merging: page content streams are copied verbatim, with no render-and-re-export round trip, so the following holds:

• Bookmarks are lost: the merged file has no outline tree. Rebuild the table of contents with PDF page management afterwards if you need one.
• No deduplication: identical pages in two source files are both kept as-is; the tool never guesses about duplicate content.
• The list decides the order: alphabetical file names have no effect — the final order is always the on-page list order.

Accepted input

standard PDF files (mixed page sizes and orientations included)

Encrypted files

PDFs with an open password must be unlocked first; the output does not inherit encryption

Damaged files

unparseable files show an error in the list and need repairing first

Size ceiling

very large files and page counts are limited by available browser memory; for hundreds of MB, merge in batches

What it does not do

no compression, no deduplication, no bookmark rebuilding, no OCR

• Combine several scans into one PDF
• Merge handouts or reports in chapter order
• Attach exhibits to a contract PDF
• Reorder several PDFs before one final export

PDF files are processed locally in the browser by default; document contents do not need to be uploaded. Data is released from memory when you close the page.

• PDF file structure basics (an on-site guide, not yet published) — the object model, page tree and incremental updates, and how they decide what merging, splitting and compression can preserve.
• ISO 32000-1:2008, Document management — Portable document format — Part 1: PDF 1.7, §7.5 file structure, §7.7.3 page tree, §7.5.7 object streams, §12.3 document-level navigation (bookmarks). Adobe public mirror: https://opensource.adobe.com/dcs/pdf/PDF32000_2008.pdf（访问日期：2026-09-08）.
• ISO 32000-2:2020 — PDF 2.0 specification（访问日期：2026-09-08） — the current definition of the page tree and document object structure.
• pdf-lib documentation, "Copy Pages" and the PDFDocument.copyPages API: https://pdf-lib.js.org/（访问日期：2026-09-08）.
• Mozilla pdf.js, the in-browser PDF renderer (used by this tool for thumbnails and password checks): https://mozilla.github.io/pdf.js/（访问日期：2026-09-08）.
• MDN Web Docs, File API overview (how files are read into browser memory): https://developer.mozilla.org/en-US/docs/Web/API/File_API（访问日期：2026-09-08）.