Merge and split PDF files

Pages are copied, not re-rendered

A PDF page is a set of drawing instructions — place this text in this font here, draw this line, show this image — rather than a picture. Merging and splitting here copy those instructions from the source file into a new one without redrawing anything. Text stays as text you can select, search, and copy; vector drawings stay sharp at any zoom; embedded images keep their original data.

We checked this rather than assuming it. After merging two test documents, the text on the first page of each was extracted back out of the merged file: both came back as real, selectable text (confirmed for both documents). That is the difference from the PDF to image converter, which deliberately turns each page into a 144 DPI picture — useful for slides and previews, but the text is no longer text.

What happened to file size when we measured it

We printed two report-style documents from Chromium — 6 and 4 pages, with headings, tables, embedded fonts, bookmarks, and accessibility tags — then ran them through the same merge and split code this page uses.

Why the merged file came out smaller

A merged file smaller than its inputs looks like something was thrown away, so we isolated the cause. Re-saving source A on its own, with no pages changed, already took it from 143 KB to 71 KB: the library writes the PDF's internal objects in a compressed form that the original file did not use. The two re-saved sources add up to 134 KB, and the merged file is 110 KB. That remaining gap is the document-level extras described below, which a merge does not carry over. The page content itself is unchanged.

Why splitting every page made the total bigger

The 6 single-page files averaged 49 KB each, about 2.1× the original in total. In the original, all pages share one copy of each embedded font. Once every page is a separate document, each file has to carry its own copy of the fonts it uses. Extracting a range behaves the same way on a smaller scale: three pages of six kept 37% of the size rather than half, because the fonts came along whole. If the goal is a smaller file to email, splitting is not the way to get it.

Measured on 2026-10-12 with Chromium 151.0.7922.34 and pdf-lib 1.17.1. Your documents will produce different absolute numbers; the direction of each effect is what transfers.

What does not carry over

Some parts of a PDF belong to the document as a whole rather than to any page. The new file is built from pages only, so these are left behind:

• Bookmarks. Both test sources had a bookmark panel; the merged file had none (measured). Long reports that rely on bookmarks for navigation will need them added again in a PDF editor.
• Accessibility tags. Both sources were tagged PDFs, the structure screen readers use to read headings, lists, and tables in order. The merged file was not (measured). If a document has to meet an accessibility requirement, merge it in software that preserves tags.
• Digital signatures. A merged or extracted file is a new document, so it cannot carry a valid signature from the original. Keep the signed original for anything that has to be verified.

Document title and author fields are also not copied from the sources. Comments and sticky notes are different: they are attached to pages, so they do travel with them — our PDF metadata test shows a reviewer's name and remark surviving a merge, an extraction and a split. For fillable forms, open the result and check the fields still behave as expected before relying on it.

Writing page ranges

In Split mode, the pages box accepts single pages and ranges separated by commas. Pages are numbered from 1, as in any PDF viewer.

• 1-3, 5 — pages 1, 2, 3 and 5.
• 5, 1 — page 5 followed by page 1. The new PDF follows the order you type, so this is also a quick way to reorder a few pages.
• 2, 2 — page 2 twice. Repeats are kept, which is handy for duplicating a cover sheet.

A range must run from the smaller number to the larger one, and a page beyond the end of the document is rejected with a message naming the last page, rather than being silently skipped.

Which BravoConvert tool to use

• Join PDFs, remove pages, or reorder a few pages while keeping text selectable: this tool.
• Turn PDF pages into JPG or PNG pictures for slides, social posts, or thumbnails: the PDF to image converter.
• Turn photos or screenshots into a PDF: the image converter with PDF selected as the output.

Limits and private processing

Each PDF can be up to 50 MB, and a merge can take up to 30 files with a combined size of 150 MB. The whole job happens in your browser's memory, so these limits exist to keep phones and older laptops from running out of it partway through. Password-protected and encrypted PDFs cannot be opened; remove the protection in the application that created the file first.

Troubleshooting

• “Not a PDF file.” The file is another format with a .pdf name, or the browser did not report it as a PDF. Open it in a PDF viewer to confirm.
• “Password-protected or encrypted.” Remove the password in the original application, save a copy, and add that copy.
• A large merge stops responding. Merge in smaller groups, then merge the results, and close other heavy tabs.
• The merged file has no bookmarks. Expected; see “What does not carry over” above.

If a valid PDF consistently fails, see the FAQ or use the contact page with your browser, device, and the error shown — but do not send the document itself.