Split PDF by page range

A page range is just a start page and an end page, inclusive. '5-12' means 'give me pages 5, 6, 7, 8, 9, 10, 11 and 12 as a new PDF'. Page numbers are 1-indexed, exactly as they appear in the page counter at the bottom of most PDF readers.

If the range goes past the last page, the tool simply stops at the end of the document. You can also pick non-contiguous pages — see the Extract Pages page for that.

Sending a single chapter from a long report instead of the whole file. Archiving the relevant section of a contract while leaving the full version on disk. Pulling the abstract and introduction out of an academic paper for a literature review. Extracting one tax form from a multi-form bundle.

The output is a new PDF containing exactly the pages you asked for, in their original order, at their original quality, with no watermark or metadata changes. The original document is untouched.

The split happens in your browser. The PDF doesn't get uploaded to a server — pages are extracted locally using a WebAssembly PDF engine.

A range is two page numbers separated by a hyphen: 1-3 means pages one, two and three combined into a single output PDF. A single number on its own — 5 — is its own one-page output. Combine ranges and single pages with commas: 1-3, 5, 7-9 produces three output PDFs. Whitespace inside the input is ignored, so '1 - 3 , 5 ' and '1-3,5' are read the same way.

Ranges may overlap. 1-5, 3-7 is valid: the first output contains pages 1 to 5, the second contains pages 3 to 7. Pages outside the document raise an error rather than being silently truncated.

If the document has a natural structure — chapters, sections, signed parties — range splitting matches that structure cleanly. You define each chunk once, run the split once, and get one file per chunk. Compared to bursting (which gives you one file per page) or extracting (which gives you a single multi-page file from scattered pages), range splitting is the right tool when the chunks are contiguous.

Pull pages 1-12 out of a 90-page proposal to send as a teaser, while keeping the full version internally. Split a 200-page scanned bundle into individual ten-page invoices. Extract just the appendix (pages 75-90) from a research paper to share with a reviewer. The point of the tool is that any of these takes about ten seconds once you've typed the range.

Range splitting runs in your browser's WebAssembly runtime. The PDF is parsed locally, the page ranges are sliced into new documents in memory, and each output file is handed to your browser as a download. No document bytes ever reach a PDF Pelican server, which matters when the reason you're splitting is to hold back the parts a third party shouldn't see.

A range is two page numbers separated by a hyphen: 1-3 means pages one, two and three combined into a single output PDF. A single number on its own — 5 — is its own one-page output. Combine ranges and single pages with commas: 1-3, 5, 7-9 produces three output PDFs. Whitespace inside the input is ignored, so '1 - 3 , 5 ' and '1-3,5' are read the same way.

Ranges may overlap. 1-5, 3-7 is valid: the first output contains pages 1 to 5, the second contains pages 3 to 7. Pages outside the document raise an error rather than being silently truncated.

If the document has a natural structure — chapters, sections, signed parties — range splitting matches that structure cleanly. You define each chunk once, run the split once, and get one file per chunk. Compared to bursting (which gives you one file per page) or extracting (which gives you a single multi-page file from scattered pages), range splitting is the right tool when the chunks are contiguous.

Pull pages 1-12 out of a 90-page proposal to send as a teaser, while keeping the full version internally. Split a 200-page scanned bundle into individual ten-page invoices. Extract just the appendix (pages 75-90) from a research paper to share with a reviewer. The point of the tool is that any of these takes about ten seconds once you've typed the range.

Range splitting runs in your browser's WebAssembly runtime. The PDF is parsed locally, the page ranges are sliced into new documents in memory, and each output file is handed to your browser as a download. No document bytes ever reach a PDF Pelican server, which matters when the reason you're splitting is to hold back the parts a third party shouldn't see.