Images in, characters out — nothing else. Recognition happens entirely on your computer; your images never leave it.
It reads Japanese (from modern print to pre-modern cursive, kuzushiji), Chinese (classical and modern), English and Tibetan. Choose the recognition engine to match the material.

News
- 16 Oct 2026 The compare screen now highlights where engines disagree (with video)
- 16 Oct 2026 PP-OCRv6 added as a recognition engine
Download
| Windows | Microsoft Store (free; the Store signs it, so no SmartScreen warning) |
| macOS | Releases (signed and notarised .dmg; macOS 12 or later) |
| iPhone / iPad | App Store (free; iOS 17 or later) |
The iPhone / iPad version is for taking a photo and reading it on the spot. Its engines are NDL Kotenseki OCR-Lite, NDLOCR-Lite, PP-OCRv6 and Apple Vision — partly different from the desktop version (no PaddleOCR-VL).
What’s new in each version is in the changelog. The Mac version doesn’t update itself, so check it now and then.
Nothing else to install. llama-server, which runs the PaddleOCR-VL model,
is bundled with the application; recognition models are fetched on first use.
Watch
Four short videos play in a row: installing, basic use, more ways to use it, and using it from the TEI/IIIF editor. The narration and captions are in Japanese.
What it does
- Input: image files, the clipboard, a folder of page images, or a IIIF manifest URL
- Output: plain text (
.txt) or TEI/XML (.xml) - Compare: run two engines over the same page and read the results side by side, with the characters where they disagree highlighted
- Japanese and English interface, light and dark
Try it without installing
IIIF Kasane also opens images from your computer in the browser and reads their text. Use it on a computer where you can’t install apps, or to try things out first. The images are not sent anywhere there either. It has fewer engines than Local OCR: NDLOCR-Lite, NDL Kotenseki OCR-Lite, PP-OCR and a few others.
Links
- How to use (in Japanese)
- Source code (MIT)
- Privacy policy
Credits
- Satoru Nakamura (The University of Tokyo)
- Tensho Miyazaki (Tsurumi University; proposed and evaluated PaddleOCR-VL)
- Yoshimitsu Aitani (Toyo Bunko; testing and feature proposals)
Contact: nakamura.satoru@mail.u-tokyo.ac.jp