The open datasets Kumi is built on, and the licenses we honor.
Kumi is built with the help of several open Japanese-language datasets maintained by dedicated projects and volunteers. We gratefully acknowledge them below. Each dataset remains the property of its owners and is used under its own license. Our own systems, software, enrichments, and original content are proprietary to Kumiai Inc.
We reproduce these datasets without modifying the underlying entries. Our additions (such as learning-priority scores, romanized search forms, lesson structure, and our own generated example sentences) are stored separately and do not alter the source data. Because we reproduce this data rather than create adaptations of it, our own systems, enrichments, and content are not subject to any ShareAlike or copyleft terms. Each source dataset remains available from its original project, linked below, under its own license.
Kumi uses the JMdict and JMnedict dictionary files. These files are the property of the Electronic Dictionary Research and Development Group, and are used in conformance with the Group's licence. Vocabulary entries, readings, and priority tags are derived from this data.
Kumi uses the KANJIDIC2 kanji dictionary file, the property of the Electronic Dictionary Research and Development Group, used in conformance with the Group's licence. Kanji meanings, readings, stroke counts, and frequency data are derived from this source.
Kanji and kana stroke-order data and component decompositions are from the KanjiVG project by Ulrich Apel, released under the Creative Commons Attribution-ShareAlike 3.0 licence. The stroke data is reproduced without modification.
Pitch-accent data is derived from the Kanjium project, released under the Creative Commons Attribution-ShareAlike 4.0 licence.
JLPT level groupings are informed by the JLPT resource lists compiled by Jonathan Waller, used under a Creative Commons Attribution licence.
If you believe a data source is missing from this page or attributed incorrectly, please let us know at legal@kumi.app and we will correct it promptly.