Split a PDF by page ranges in your browser — each range becomes its own file, pages copied losslessly. Processed entirely on your device, never uploaded. Runs entirely in your browser — your numbers never leave this device.
No PDF yet. Choose one above, then type the page ranges you want as separate files — each comma-separated range becomes one PDF.
Your file is processed entirely in your browser and never uploaded: the PDF is read into memory on your own device, the pages named by your ranges are copied into new documents, and each range is offered back as its own download. Nothing is sent to a server at any stage — the tool has no server side and works offline once the page has loaded. Like the other structural tools in this category, splitting copies pages as objects rather than re-rendering them, so extracted pages keep their text layer, vectors, page size and rotation exactly.
Ranges are written the way people say them: single pages and hyphenated spans, separated by commas — “1-3, 7” means pages 1 to 3 in one file and page 7 alone in a second. The count for any span is b − a + 1, because both endpoints are included: 1-3 is 3 pages, not 2. Ranges may repeat or overlap (nothing stops you extracting page 5 into two different outputs) and they need not cover the whole document — pages named in no range are simply left out of every output, which is also how this tool doubles as a page extractor.
Splitting is the natural partner of merging. Split a document into sections, reorder or discard pieces, then use Merge PDF to assemble exactly the document you want; or split a long report into chapter files for separate sharing. What splitting does not do is divide one page into pieces — a page is the smallest unit a PDF has — and it does not shrink anything: the outputs together hold copies of the named pages, and their combined size is usually close to the original's.
No. The file is opened, copied from and downloaded entirely inside your browser; it never leaves your device, is never stored or logged, and the tool works with the network off.
No. Pages are copied structurally, exactly as in Merge PDF: text stays selectable, vectors stay vector, and each page keeps its size and rotation. Nothing is re-rendered.
A range naming pages the document does not have cannot be fulfilled — a 10-page file has no page 11. Check the page count the tool reports for your file and keep every range inside 1 to that count.
Yes. Ranges are independent: listing “1-5” and “5-9” produces two files that both contain page 5. Overlap is allowed because each range is copied separately.
They are two views of the same job. Split names the pages you want to keep, one output per range; Delete names the pages you want to remove and returns a single file with the rest. For one continuous remainder, Delete is shorter to specify; for several separate outputs, Split is the tool.
More in PDF Tools · Browse all tools