Open Source Kanji Data

KanjiTest.Online publishes all of its JLPT kanji, grammar, and radical data as open datasets. You are free to use these datasets in your own projects — apps, flashcards, study tools, research, or anything else.

JSON API Endpoints

All endpoints return structured JSON and are updated whenever the site data changes.

Data Format

Each kanji entry includes:

• kanji — the character itself
• kun — kun'yomi readings
• on — on'yomi readings
• meanings — English meanings
• strokes — stroke count
• examples — vocabulary words with readings, English translations, and full sentences
• level — JLPT level (n5–n1)

How to Use

Fetch the data directly in your JavaScript, Python, or any other language:

// JavaScript
const kanji = await fetch('https://www.kanjitest.online/api/decks.json').then(r => r.json())
console.log(kanji.n5[0]) // first N5 kanji

# Python
import requests
data = requests.get('https://www.kanjitest.online/api/decks.json').json()
print(data['n5'][0]['kanji'], data['n5'][0]['meanings'])

LLMs.txt

A machine-readable summary of the site is available for AI assistants and large language models:llms.txt. This file provides a structured overview of all content sections, APIs, and navigation paths.

License

All datasets are released under theMIT License. You may use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the data without restriction. Attribution is appreciated but not required.