Conversion cases with evidence

Each case is a real PDF from our licensed or public-domain corpus, converted with the same engine you use, with the results of every engine we measured, including where ours loses.

Updated: Nov 3, 2026

• Two-column PDF to EPUB: an academic paper

  OLMo paper (arXiv 2402.00838)

• Three-column PDF to EPUB: the Federal Register

  Federal Register document 2026-22443

• PDF footnotes to EPUB notes: an academic paper

  QLoRA paper (arXiv 2305.14314)

• Removing repeated headers: the 9/11 Commission Report

  The 9/11 Commission Report (GPO)

• Chapters and table of contents from a PDF book

  Calculus textbook typeset in LaTeX (Project Gutenberg #33283)

• Arabic PDF to EPUB: right-to-left text

  Wikipedia (Arabic): One Thousand and One Nights

• Japanese PDF to EPUB

  Wikipedia (Japanese): The Tale of Genji

See the full benchmark