Kreuzberg is a high-performance document intelligence library with a Rust core and native Ruby bindings via Magnus. Extract text, metadata, and structured data from 75+ file formats including PDF, DOCX, PPTX, XLSX, HTML, RTF, images (with OCR), email, archives, and more. Features async/sync APIs, text chunking, language detection, and keyword extraction.

Required Ruby Version

>= 3.2.0, < 5.0

Authors

Na'aman Hirschfeld

Versions

  1. 5.0.0.pre.rc.35

    1. source

      1. 5.0.0.pre.rc.35 source June 24, 2026 (44.6 MB)
  2. 5.0.0.pre.rc.34

    1. source

      1. 5.0.0.pre.rc.34 source June 24, 2026 (44.8 MB)
  3. 5.0.0.pre.rc.32

    1. source

      1. 5.0.0.pre.rc.32 source June 23, 2026 (43.7 MB)
  4. 5.0.0.pre.rc.30

    1. source

      1. 5.0.0.pre.rc.30 source June 23, 2026 (44.7 MB)
  5. 5.0.0.pre.rc.29

    1. source

      1. 5.0.0.pre.rc.29 source June 22, 2026 (44.7 MB)
Show all versions (243 total)

Pushed by

GitHub

SHA 256 checksum

← Previous version Next version →