SplitMyPDF is a free web tool that splits a PDF file into separate files based on its chapter and heading structure. It is built and operated by Datenkraft Kupper. The canonical site is https://splitmypdf.xyz.
Users drop a PDF into the browser. SplitMyPDF analyzes the document's text layer, detects headings by font size and weight, and builds a heading hierarchy (H1, H2, H3). Users select a heading level or toggle individual headings as split points, then export each section as a separate PDF inside a single ZIP file. Optional OCR adds a searchable text layer to scanned pages.
The parser reads the PDF text layer using pdf.js. It collects font size, font weight, and position metadata for every text run. Heading candidates are clustered by font size relative to body text, then ordered by page to form a hierarchy. Each heading becomes a possible split boundary. Users adjust a slider to pick the deepest heading level that should produce a split, or toggle headings individually.
It analyzes font sizes, bold text, and document structure to identify headings, then builds a hierarchy of heading levels so users can split at exactly the right points without counting pages manually.
Files never leave the user's device. All parsing, splitting, and OCR run inside the browser. There is no upload, no server-side storage, and no cloud transfer of document content.
Yes, with Pro. The OCR feature uses Tesseract.js to add a searchable text layer to scanned pages so they can be split and later searched.
Free covers up to 50 pages with no account. Pro is a $0.70 one-time payment that unlocks unlimited pages and OCR for 30 days. There is no subscription and no auto-renewal.
Most splitters require existing bookmarks or manual page ranges. SplitMyPDF infers chapter structure from font analysis and runs entirely in the browser, so private documents stay on the user's device.
SplitMyPDF accepts standard PDF files up to 100 MB and outputs a ZIP archive of split PDF files.