Split PDF

Range syntax explained, with worked examples

The range box understands a tiny grammar on purpose: numbers, dashes and commas. A bare number such as 7 means "just page seven". A pair joined by a dash, like 4-9, means "pages four through nine, inclusive". Join as many segments as you like with commas and the engine extracts the union — so 1-3,4,5-9 means pages one through three, plus page four, plus pages five through nine, which for a ten-page document is everything except page ten.

Two safety rules apply as you type. First, the moment the file loads the tool counts its pages and shows the total next to the subtitle, so you can sanity-check the range before clicking. Second, every segment is validated against that total: asking for 15-20 on a 12-page document throws an error naming the document length instead of silently exporting garbage. Backwards ranges like 9-4 are also normalized to 4-9.

• Extract a cover plus appendix: 1, 12-14 on a 14-page deck.
• Send only the middle chapter of an internal report: 5-9.
• Split a signed form into two halves for two reviewers: use ZIP mode with 1-3,4-6 and you get two three-page PDFs in one archive.
• Whitespace around segments is ignored, so "1 - 3, 5" parses the same as "1-3,5".

Single-PDF output vs ZIP output, and what to do with the result

Single mode merges all your requested pages into one new PDF, keeping the reading order — pages sorted ascending, even if you typed the range out of order. This is what you want when you are emailing the relevant sections of a long deck or removing an appendix nobody needs. ZIP mode instead runs the splitter once per comma-separated segment and bundles each result with JSZip: part-1 covers your first segment, part-2 the second, and so on, with the segment label folded into the filename.

Splitting is often the first move in a cleanup pass rather than the last. Pull the pages you need, add page numbers if the source deck was unnumbered, merge extra documents you dropped on top, and compress the final file before it goes out. If the purpose is to reuse a diagram or quote a paragraph, the PDF to image converter exports the extracted pages as PNG or JPEG instead.

How to find the right page numbers before typing them

The most common splitting mistake is trusting the page numbers printed on the document, not the page numbers in the PDF reader. Word-processed decks often have a cover, a title page and a table of contents before the body starts, so the printed "page 1" is actually the fourth physical page of the file. Before you type a range, jump to the target page in your browser PDF viewer and check the thumbnail sidebar: that number is what this tool counts from.

A few practical tips follow from that. If the document has bookmarks, use them to navigate to the right section and read off the physical page number from the viewer. For duplex printing remember that a two-page spread shows one even-numbered and one odd-numbered side together, so "left page" is usually an even number. Rotated pages keep their rotation when extracted, so sideways landscape pages come out sideways — that is the original layout, not a defect. Encrypted files cannot be opened by the local engine and are rejected up front, so unlock the document in your reader first and then split the unlocked copy.

Frequently Asked Questions