Compare PDF text
Compare text changes page by page between two PDFs and save a difference report. Images, formatting and scanned-page OCR are not compared.
PDFs to compare
Editable public example without files
Only text at matching page numbers is compared. Extracted items are separated by spaces or line breaks and trailing page whitespace is removed. One changed span is shown after removing the common prefix and suffix; it may include unchanged text between edits. Inserted pages and reflow can enlarge differences. Images, formatting and signatures are not compared; this does not establish document equivalence.
Files and extracted text stay in browser memory and are cleared on reload or navigation. Names and contents are excluded from analytics.
Comparison by page
No comparison yet.
Pages without text may be scanned, image-only or blank and were not assessed. Matching extracted text does not check words inside images. Review the originals or use an OCR tool.
Instructions and example
The public original contains Order 1001 and Total 100.00; the revision contains Order 1001 and Total 120.00. Comparing shows the amount change. For real PDFs, extracted strings are paired by page number. The text between their common prefix and suffix is shown as one changed span, so unchanged text between separate edits on the same page may also appear in that span.
Frequently asked questions
Does no text difference mean the documents are identical?
No. Only extracted text is compared. This does not assess images, fonts, layout, annotations, signatures, hidden objects or legal equivalence.
How are scanned PDFs and blank pages handled?
Pages without a text layer are marked unassessed. They are never treated as identical merely because no text was extracted, and OCR is not run. With mixed text and scanned pages, the result does not assess the entire document.
What about inserted pages or large PDFs?
Each file supports up to 20MiB, 50 pages and 200,000 extracted characters. Pages are paired by number, so inserting a page or reflowing text may cause large subsequent differences. Encrypted or damaged documents may fail.
Calculation criteria and reference materials
- Mozilla — PDF.js
Reference for extracting PDF text. This tool compares text by page number, not image or formatting equivalence.
Verification of calculation criteria: · Calculation·verification principles · Report errors