Merge PDF — Free Online Tool on Toolpile
Every tool you need, one place
PDF, Image, AI, Dev, and Business tools — free, private, no signup.
Toolpile uses cookies for analytics and ads. The tools themselves keep your files in the browser — that doesn't change. Read the full policy
PDF, Image, AI, Dev, and Business tools — free, private, no signup.
About Merge PDF
Merging PDFs sounds trivial — concatenate two files. In practice the file format is a graph of references, not a stream of pages, and "merge" can quietly break bookmarks, links, fonts, or even rotation depending on which library does it. Here's what's actually happening, and why this tool keeps it boring.
A PDF is a tree of objects: each page is an object, each font is an object, each image is an object, and a top-level catalog points at the page tree. Merging two PDFs means rewriting both catalogs into one — copying every page object plus every resource it references (fonts, images, color profiles, form fields, annotations) into a single new file and updating internal IDs so nothing collides. Done well, the output looks identical to opening both originals side by side. Done badly, you get duplicated embedded fonts (file size doubles), broken hyperlinks (annotation targets point at deleted IDs), or lost form data (AcroForm fields get orphaned).
This tool uses pdf-lib, which copies pages with their full reference chains intact and de-duplicates fonts/images by content hash. That means merging ten copies of the same PDF doesn't multiply file size 10×; identical embedded resources are stored once. It also means bookmarks and internal links from each source are preserved — a hyperlink in document A still points at page A.5, even when A.5 is now page 23 of the merged file.
Things that don't always survive a merge in any tool: digital signatures (signing covers a specific byte range; reordering invalidates it — by design), AcroForm field calculations that reference fields by absolute name (two PDFs with a "total" field will collide), and PDF/A conformance (merging a PDF/A-2 with a non-conforming PDF downgrades the result). If any of those matter, merge in a PDF-aware editor that warns you, not in a generic merger.
PDFs store rotation as a per-page metadata flag (0/90/180/270) that viewers apply at render time — the underlying content stream isn't physically rotated. When you merge a portrait scan with a landscape spreadsheet that was saved with a 90° rotation flag, the merged file preserves both flags, so each page renders correctly. But if a downstream tool reads the raw content stream and ignores rotation flags (some print drivers, OCR tools), the spreadsheet page comes out sideways. The fix is to flatten rotation before merging — the Rotate PDF tool on this site does that explicitly.
Page order is determined entirely by the order you upload files; this tool concatenates in that sequence and within each file preserves the original page order. There's no shuffling. If you need an interleaved order (page 1 of A, page 1 of B, page 2 of A, page 2 of B — a common need for scanning duplex documents in two passes), split each into single pages first, then merge in the interleaved sequence. The Split PDF tool on this site exports to one file per page for exactly this case.
Are my PDFs uploaded anywhere?
No. The merge runs entirely in your browser via pdf-lib. There is no server-side processing, no temporary upload, no analytics on file content. The only network traffic is the page itself loading. You can disconnect from the internet after the page opens and the merger still works.
Will this work with password-protected PDFs?
No — pdf-lib refuses to read encrypted streams (deliberately, because handling decryption client-side without exposing the password is risky). Run each protected PDF through the Unlock PDF tool first (which also runs locally), then merge the unlocked copies.
How big can the input files be?
Browser memory is the practical limit. Most modern laptops handle 200-500 MB total comfortably; 1 GB+ pushes Chrome's per-tab memory budget and may crash the tab. If you need to merge huge scans, split them into chunks of 5-10 files each, merge those, then merge the merged outputs.
Does merging compress the file or change quality?
Neither. Merging preserves every page's content stream byte-for-byte; image quality, font hinting, and color profiles are unchanged. The output may be slightly smaller than the sum of inputs because pdf-lib de-duplicates identical embedded fonts and images, but it never re-encodes content.
Why is the merged file larger than I expected?
Two usual causes: (1) each source embeds a different font subset of the same family, so the merged file carries multiple subsets (de-dup only works on byte-identical resources), or (2) each source contains the same image at slightly different bit-exact encodings. To shrink, run the merged output through Compress PDF, which re-encodes images and consolidates fonts.
Will hyperlinks, bookmarks, and form fields survive?
Hyperlinks (both internal cross-page and external URLs) and bookmarks are preserved because pdf-lib copies the annotation and outline trees with the page. Interactive form fields (AcroForm) are preserved per-page but field-name collisions across files will overwrite — if both PDFs have a field named "signature", the second one wins. Rename fields before merging if that matters.
Can I reorder pages after merging?
Not in this tool — output order matches input order. To reorder pages within a single PDF, use Split PDF (export to one-page files), drag to the order you want, then re-merge. For frequent rearranging, a desktop PDF editor with a thumbnail panel is faster than a chain of web tools.
Are PDF/A or signed PDFs handled correctly?
Digital signatures invalidate when content is reordered — this is by design, not a tool limitation; any merge breaks the signature. Re-sign the merged file if needed. PDF/A conformance is preserved only if every input is the same PDF/A version; mixing PDF/A-1 and PDF/A-2 (or PDF/A with regular PDF) downgrades the output to plain PDF.