awesome-japanese-nlp-resources

A curated list of resources dedicated to Python libraries, llms, dictionaries, and corpora of NLP for Japanese

• Listed information on 895 GitHub repositories
• Listed information on 450 Hugging Face repositories (models and datasets)
• Listed information on 156 multilingual libraries, models, and datasets that also support Japanese

English | 日本語 (Japanese) | 繁體中文 (Chinese) | 简体中文 (Chinese)

🎉 The latest additions

Corpus

• jmeeteval - J-MeetEval: 日本語会議タスクにおける指示追従性評価ベンチマーク

Updated on Oct 13, 2026

Claude Code & Codex Plugin

Search, discover, and track Japanese NLP resources directly from Claude Code or Codex using the awesome-japanese-nlp-resources plugin. Both tools use the same skills and the same bundled data.

Claude Code:

# Add the marketplace and install
/plugin marketplace add taishi-i/awesome-japanese-nlp-resources
/plugin install awesome-japanese-nlp-resources@awesome-japanese-nlp-resources
/reload-plugins

Codex (run in your terminal, then start a new Codex session):

codex plugin marketplace add taishi-i/awesome-japanese-nlp-resources
codex plugin add awesome-japanese-nlp-resources@awesome-japanese-nlp-resources

The plugin ships four skills:

In Codex, call the same skills with $ instead of / (e.g. $awesome-japanese-nlp-resources:search ).

All skills detect the query language and respond in kind — English by default, Japanese when the query contains Japanese characters.

search example:

/awesome-japanese-nlp-resources:search What tutorials are recommended for beginners learning Japanese NLP?

Output includes a Use-case Selection Guide table:

For full documentation, see the plugin README.

Contents

• Hugging Face
  • Models
  • Datasets
• Python library
  • Morphology analysis
  • Parsing
  • Converter
  • Preprocessor
  • Sentence splitter
  • Sentiment analysis
  • Machine translation
  • Named entity recognition
  • OCR
  • Tool for pretrained models
  • Others
• C++
  • Morphology analysis
  • Parsing
  • Others
• Rust crate
  • Morphology analysis
  • Converter
  • Search engine library
  • Others
• JavaScript
  • Morphology analysis
  • Converter
  • Others
• Go
  • Morphology analysis
  • Others
• Java
  • Morphology analysis
  • Others
• Pretrained model
  • Word2Vec
  • Transformer based models
• ChatGPT
• Dictionary and IME
• Corpus
  • Part-of-speech tagging / Named entity recognition
  • Parallel corpus
  • Dialog corpus
  • Others
• Tutorial
• Research summary
• Reference
• Contributors

Python library

Morphology analysis

Libraries that split Japanese text into words or morphemes and assign part-of-speech and base forms

• sudachi.rs - SudachiPy 0.6* and above are developed as Sudachi.rs.
• Janome - Japanese morphological analysis engine written in pure Python
• mecab-python3 - mecab-python. you can find original version here:http://taku910.github.io/mecab/
• mecab - This repository is for building Windows 64-bit MeCab binary and improving MeCab Python binding.
• fugashi - A Cython MeCab wrapper for fast, pythonic Japanese tokenization and morphological analysis.
• nagisa - A Japanese tokenizer based on recurrent neural networks
• pyknp - A Python Module for JUMAN++/KNP
• Mykytea-python - Python wrapper for KyTea
• konoha - Konoha: Simple wrapper of Japanese Tokenizers
• natto-py - natto-py combines the Python programming language with MeCab, the part-of-speech and morphological analyzer for the Japanese language.
• rakutenma-python - Rakuten MA (Python version)
• python-vaporetto - Vaporetto is a fast and lightweight pointwise prediction based tokenizer. This is a Python wrapper for Vaporetto.
• dango - An easy to use tokenizer for Japanese text, aimed at language learners and non-linguists
• rhoknp - Yet another Python binding for Juman++/KNP
• python-vibrato - Viterbi-based accelerated tokenizer (Python wrapper)
• jagger-python - Python binding for Jagger(C++ implementation of Pattern-based Japanese Morphological Analyzer)
• Mecari - Mecari (Japanese Morphological Analysis with Graph Neural Networks)

Parsing

Libraries that analyze syntactic and dependency structures of Japanese sentences

• ginza - A Japanese NLP Library using spaCy as framework based on Universal Dependencies
• cabocha - Yet Another Japanese Dependency Structure Analyzer
• UniDic2UD - Tokenizer POS-tagger Lemmatizer and Dependency-parser for modern and contemporary Japanese
• camphr - Camphr - NLP libary for creating pipeline components
• SuPar-UniDic - Tokenizer POS-tagger Lemmatizer and Dependency-parser for modern and contemporary Japanese with BERT models
• depccg - A* CCG Parser with a Supertag and Dependency Factored Model
• bertknp - A Japanese dependency parser based on BERT
• esupar - Tokenizer POS-Tagger and Dependency-parser with BERT/RoBERTa/DeBERTa models for Japanese and other languages
• yomikata - Heteronym disambiguation library using a fine-tuned BERT model.
• jdepp-python - Python binding for J.DepP(C++ implementation of Japanese Dependency Parsers)
• lightblue - A CCG parser for Japanese with DTS-representations
• natsume-simple - natsume-simpleは日本語の係り受け関係検索システム
• jdeppy - Python wrapper for J.DepP, fast Japanese Dependency Parser

Converter

Libraries that convert between character types such as kana, romaji, and full-width/half-width forms

• pykakasi - Lightweight converter from Japanese Kana-kanji sentences into Kana-Roman.
• cutlet - Japanese to romaji converter in Python
• alphabet2kana - Convert English alphabet to Katakana
• Convert-Numbers-to-Japanese - Converts Arabic numerals, or ‘western’ style numbers, to a Japanese context.
• mozcpy - Mozc for Python: Kana-Kanji converter
• jamorasep - Japanese text parser to separate Hiragana/Katakana string into morae (syllables).
• text2phoneme - 日本語文を音素列へ変換するスクリプト
• jntajis-python - A fast character conversion and transliteration library based on the scheme defined for Japan National Tax Agency (国税庁) 's
• wiredify - Convert japanese kana from ba-bi-bu-be-bo into va-vi-vu-ve-vo
• mecab-text-cleaner - Simple Python package (CLI/Python API) for getting japanese readings (yomigana) and accents using MeCab.
• pynormalizenumexp - 数量表現や時間表現の抽出・正規化を行うNormalizeNumexpのPython実装
• Jusho - Easy wrapper for the postal code data of Japan
• yurenizer - Japanese text normalizer that resolves spelling inconsistencies. （日本語表記揺れ解消ツール）
• e2k - A tool for automatic English to Katakana conversion
• alkana.py - A tool to get the katakana reading of an alphabetical string.
• englishtokanaconverter - 英語文字列をカタカナに変換するプログラム
• kanjiconv - Kanji Converter to Hiragana, Katakana, Roman alphabet.
• kanjize - Kanjize(カンジャイズ): Easy converter between Kanji-Number and Integer

Preprocessor

Libraries that normalize and clean text before analysis

• neologdn - Japanese text normalizer for mecab-neologd
• jaconv - Pure-Python Japanese character interconverter for Hiragana, Katakana, Hankaku, and Zenkaku
• mojimoji - A fast converter between Japanese hankaku and zenkaku characters
• text-cleaning - A powerful text cleaner for Japanese web texts
• HojiChar - 複数の前処理を構成して管理するテキスト前処理ツール
• utsuho - Utsuho is a Python module that facilitates bidirectional conversion between half-width katakana and full-width katakana in Japanese.
• python-habachen - Yet Another Fast Japanese String Converter
• kairyou - Quickly preprocesses Japanese text using NLP/NER from SpaCy for Japanese translation or other NLP tasks.

Sentence splitter

Libraries that automatically detect sentence boundaries and split text

• Bunkai - Sentence boundary disambiguation tool for Japanese texts (日本語文境界判定器)
• japanese-sentence-breaker - Japanese Sentence Breaker
• sengiri - Yet another sentence-level tokenizer for the Japanese text
• budoux - Standalone. Small. Language-neutral. BudouX is the successor to Budou, the machine learning powered line break organizer tool.
• ja_sentence_segmenter - japanese sentence segmentation library for python
• hasami - A tool to perform sentence segmentation on Japanese text
• kuzukiri - Japanese Text Segmenter for Python written in Rust
• ja-senter-benchmark - Comparison of Japanese Sentence Segmentation Tools
• fast-bunkai - Japanese sentence splitting(日本語文境界判定器), 40–250× faster via a Rust-accelerated Python library with near-perfect API compatibility with megagonlabs/bunkai.

Sentiment analysis

Libraries that detect emotions or polarity in text

• oseti - Dictionary based Sentiment Analysis for Japanese
• negapoji - Japanese negative positive classification.日本語文書のネガポジを判定。
• pymlask - Emotion analyzer for Japanese text
• asari - Japanese sentiment analyzer implemented in Python.
• kotobacore - Japanese semantic understanding engine for LLM, RAG, SNS analysis, and AI agents ? dependency-free tokenizer + Plutchik emotion + intent + RAG keywords.

Machine translation

Libraries that automatically translate text between languages

• jparacrawl-finetune - An example usage of JParaCrawl pre-trained Neural Machine Translation (NMT) models.
• JASS - JASS: Japanese-specific Sequence to Sequence Pre-training for Neural Machine Translation (LREC2020) & Linguistically Driven Multi-Task Pre-Training for Low-Resource Neural Machine Translation (ACM TALLIP)
• PheMT - A phenomenon-wise evaluation dataset for Japanese-English machine translation robustness. The dataset is based on the MTNT dataset, with additional annotations of four linguistic phenomena; Proper Noun, Abbreviated Noun, Colloquial Expression, and Variant. COLING 2020.
• VISA - An ambiguous subtitles dataset for visual scene-aware machine translation
• plamo-translate-cli - A command-line interface for translation using the plamo-2-translate model with local execution.

Named entity recognition

Libraries that extract names of people, places, and organizations from text

• namaco - Character Based Named Entity Recognition.
• entitypedia - Entitypedia is an Extended Named Entity Dictionary from Wikipedia.
• noyaki - Converts character span label information to tokenized text-based label information.
• bert-japanese-ner-finetuning - Code to perform finetuning of the BERT model. BERTモデルのファインチューニングで固有表現抽出用タスクのモデルを作成・使用するサンプルです
• joint-information-extraction-hs - 詳細なアノテーション基準に基づく症例報告コーパスからの固有表現及び関係の抽出精度の推論を行うコード
• pygeonlp - pygeonlp, A python module for geotagging Japanese texts.
• bert-ner-japanese - BERTによる日本語固有表現抽出のファインチューニング用プログラム
• huggingface-finetune-japanese - Examples to finetune encoder-only and encoder-decoder transformers for Japanese language (Hugging Face) Resources
• novelanalysisbyner - BERTのfine-tuningによる固有表現抽出

OCR

Libraries that recognize and extract text from images

• Manga OCR - About Optical character recognition for Japanese text, with the main focus being Japanese manga
• mokuro - Read Japanese manga inside browser with selectable text.
• handwritten-japanese-ocr - Handwritten Japanese OCR demo using touch panel to draw the input text using Intel OpenVINO toolkit
• OCR_Japanease - 日本語OCR
• ndlocr_cli - NDLOCRのアプリケーション
• donut - Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022
• JMTrans - manga translator - get japanese manga from url to translate manga image
• Kindai-OCR - OCR system for recognizing modern Japanese magazines
• text_recognition - NDLOCR用テキスト認識モジュール
• Poricom - Optical character recognition in manga images. Manga OCR desktop application
• owocr - Optical character recognition for Japanese text
• yomitoku - Yomitoku is an AI-powered document image analysis package designed specifically for the Japanese language.
• findtextcenternet - Japanese OCR with CenterNet
• simple-ocr-for-manga - A simple OCR for manga (Japanese traditional and Japanese vertical)
• jp-ocr-evaluation - 日本語の文章画像に対するOCRの性能を評価
• paddleocr-vl-sft-for-japanese-manga-on-rtx-3060 - Fine-tune PaddleOCR-VL on the Manga109s dataset for Japanese manga text recognition. The base model struggles with vertical Japanese text reading order in manga. After fine-tuning, the model correctly handles manga-specific text layouts.
• MangaOCR - A lightweight OCR model for Japanese text, especially in Manga
• meikiocr - high-speed, high-accuracy, local ocr for japanese video games
• meikipop - universal japanese ocr popup dictionary for windows, linux and macos

Tool for pretrained models

Libraries that utilize pretrained models to improve accuracy and efficiency

• JGLUE - JGLUE: Japanese General Language Understanding Evaluation
• ginza-transformers - Use custom tokenizers in spacy-transformers
• t5_japanese_dialogue_generation - T5による会話生成
• japanese_text_classification - To investigate various DNN text classifiers including MLP, CNN, RNN, BERT approaches.
• Japanese-BERT-Sentiment-Analyzer - Deploying sentiment analysis server with FastAPI and BERT
• jmlm_scoring - Masked Language Model-based Scoring for Japanese and Vietnamese
• allennlp-shiba-model - AllenNLP integration for Shiba: Japanese CANINE model
• evaluate_japanese_w2v - script to evaluate pre-trained Japanese word2vec model on Japanese similarity dataset
• gector-ja - BERT-based GEC tagging for Japanese
• Japanese-BPEEncoder - Japanese-BPEEncoder
• Japanese-BPEEncoder_V2 - Japanese-BPEEncoder Version 2
• transformer-copy - 日本語文法誤り訂正ツール
• nagisa_bert - A BERT model for nagisa
• JGLUE-benchmark - Training and evaluation scripts for JGLUE, a Japanese language understanding benchmark
• jptranstokenizer - Japanese Tokenizer for transformers library
• jp-stable - JP Language Model Evaluation Harness
• compare-ja-tokenizer - How do different tokenizers perform on downstream tasks in scriptio continua languages?: A case study in Japanese-ACL SRW 2023
• lm-evaluation-harness-jp-stable - A framework for few-shot evaluation of autoregressive language models.
• llm-lora-classification - llm-lora-classification
• rinna_gpt-neox_ggml-lora - The repository contains scripts and merge scripts that have been modified to adapt an Alpaca-Lora adapter for LoRA tuning when assuming the use of the “rinna/japanese-gpt-neox…” [gpt-neox] model converted to ggml.
• japanese-llm-roleplay-benchmark - このリポジトリは日本語LLMのキャラクターロールプレイに関する性能を評価するために作成しました。
• japanese-llm-ranking - This repository supports YuzuAI’s Rakuda leaderboard of Japanese LLMs, which is a Japanese-focused analogue of LMSYS’ Vicuna eval.
• llm-jp-eval - このツールは、複数のデータセットを横断して日本語の大規模言語モデルを自動評価するものです．
• llm-jp-sft - This repository contains the code for supervised fine-tuning of LLM-jp models.
• llm-jp-tokenizer - LLM勉強会（LLM-jp）で開発しているLLM用のトークナイザー関連をまとめたリポジトリです．
• japanese-lm-fin-harness - Japanese Language Model Financial Evaluation Harness
• ja-vicuna-qa-benchmark - Japanese Vicuna QA Benchmark
• swallow-evaluation - Swallowプロジェクト 大規模言語モデル 評価スクリプト
• swallow-evaluation-instruct - Swallowプロジェクト 事後学習ずみ大規模言語モデル 評価フレームワーク
• pretrained_doc2vec_ja - pretrained doc2vec models on Japanese Wikipedia
• pl-bert-ja - A repository of Japanese Phoneme-Level BERT

Others

General-purpose tools supporting Japanese language processing

C++

Morphology analysis

High-performance libraries for Japanese morphological analysis

• mecab - Yet another Japanese morphological analyzer
• jumanpp - Juman++ (a Morphological Analyzer Toolkit)
• kytea - The Kyoto Text Analysis Toolkit for word segmentation and pronunciation estimation, etc.
• juman - Japanese Morphological Analysis System JUMAN

Parsing

Libraries for dependency and syntactic parsing of Japanese sentences

• cabocha - Yet Another Japanese Dependency Structure Analyzer
• knp - A Japanese Parser

Others

Other Japanese NLP and text processing libraries

Rust crate

Morphology analysis

Fast Japanese morphological analysis crates written in Rust

• lindera - A morphological analysis library.
• vaporetto - Vaporetto: Very Accelerated POintwise pREdicTion based TOkenizer
• goya - Japanese Morphological Analysis written in Rust
• vibrato - vibrato: Viterbi-based accelerated tokenizer
• yoin - A Japanese Morphological Analyzer written in pure Rust
• mecab-rs - Safe Rust bindings for mecab a part-of-speech and morphological analyzer library
• awabi - A morphological analyzer using mecab dictionary
• kanpyo - Japanese Morphological Analyzer written in Rust
• mecrab - A pure Rust implementation of a morphological analyzer compatible with MeCab dictionaries (IPADIC format).

Converter

Crates for script and character conversion in Japanese text

• wana_kana_rust - Utility library for checking and converting between Japanese characters - Hiragana, Katakana - and Romaji
• unicode-jp-rs - A Rust library to convert Japanese Half-width-kana[半角ｶﾅ] and Wide-alphanumeric[全角英数] into normal ones
• kana - [Mirror] CLI program for transliterating romaji text to either hiragana or katakana
• kanaria - このライブラリは、ひらがな・カタカナ、半角・全角の相互変換や判別を始めとした機能を提供します。
• japanese-address-parser - 日本の住所を都道府県/市区町村/町名/その他に分割するライブラリです
• yosina - Yosina is a transliteration library deals with the letters and symbols used in Japanese writing.
• mojimoji-rs - Rust implementation of a fast converter between Japanese hankaku and zenkaku characters, mojimoji.
• haqumei - A Japanese Grapheme-to-Phoneme (G2P) library.
• ja-furigana - 日本語フリガナ (ルビ) を扱う Rust 製ライブラリ + ローカル HTTP サーバー。ルールはすべてデータ駆動 (TOML)。
• jpnorm - 日本語テキスト正規化ライブラリ (Rust core + Python)。neologdn 互換・用途別プリセット・URL 保護・カスタム辞書 / Fast, configurable Japanese text normalization

Search engine library

Libraries for Japanese full-text search and indexing

• lindera-tantivy - Lindera tokenizer for Tantivy.
• tantivy-vibrato - A Tantivy tokenizer using Vibrato.
• sqlite-vaporetto - SQLite FTS5 extension for fast Japanese full-text search with 🛥Vaporetto / Vaporetto による高速な日本語全文検索を SQLite FTS5 で実現する拡張機能
• duckdb-vaporetto - DuckDB extension for Japanese full-text search with 🛥Vaporetto / Vaporetto による DuckDB + 日本語全文検索拡張機能
• lindera-sqlite - Lindera for SQLite FTS5 extention

Others

Supplementary crates for Japanese text and IME processing

JavaScript

Morphology analysis

Japanese morphological analysis libraries for browser and Node.js

• kuromoji.js - JavaScript implementation of Japanese morphological analyzer
• rakutenma - Rakuten MA - morphological analyzer (word segmentor + PoS Tagger) for Chinese and Japanese written purely in JavaScript.
• node-mecab-ya - Yet another mecab wrapper for nodejs
• juman-bin - a User-Extensible Morphological Analyzer for Japanese. 日本語形態素解析システム
• node-mecab-async - Asynchronous japanese morphological analyser using MeCab.

Converter

Libraries for converting Japanese scripts and readings

• kuroshiro - Japanese language library for converting Japanese sentence to Hiragana, Katakana or Romaji with furigana and okurigana modes supported.
• kuroshiro-analyzer-kuromoji - Kuromoji morphological analyzer for kuroshiro.
• hepburn - Node.js module for converting Japanese Hiragana and Katakana script to, and from, Romaji using Hepburn romanisation
• japanese-numerals-to-number - Converts Japanese Numerals into number
• jslingua - Javascript libraries to process text: Arabic, Japanese, etc.
• WanaKana - Javascript library for detecting and transliterating Hiragana Katakana Romaji
• node-romaji-name - Normalize and fix common issues with Romaji-based Japanese names.
• kyujitai.js - Utility collections for making Japanese text old-fashioned
• normalize-japanese-addresses - オープンソースの住所正規化ライブラリ。
• jaconv - 日本語文字変換ライブラリ (javascript)
• romaji-conv - Convert romaji into hiragana
• japanese-addresses-v2 - 全国の住所データAPI
• jptext-to-emoji - テキストの単語を絵文字に変換する
• japanese.js - Util collection for Japanese text processing. Hiraganize, Katakanize, and Romanize.
• genshijin - About genshijin 原始人 🗿| Claude Code / Codex等AIエージェント 向け超圧縮コミュニケーションスキル。caveman の日本語版をベースに、日本語特有の冗長表現に最適化。

Others

Other libraries for Japanese NLP in JavaScript

Go

Morphology analysis

Lightweight Japanese morphological analysis libraries in Go

• kagome - Self-contained Japanese Morphological Analyzer written in pure Go

Others

Additional Go-based Japanese text processing libraries

Java

Morphology analysis

Japanese morphological analysis and dictionary management libraries

• kuromoji - Kuromoji is a self-contained and very easy to use Japanese morphological analyzer designed for search
• Sudachi - A Japanese Tokenizer for Business
• SudachiDict - A lexicon for Sudachi
• meval - 形態素解析器性能評価システム MevAL

Others

Java libraries for Japanese NLP and OCR

Pretrained model

Word2Vec

Models that convert words into numeric vectors to capture semantic similarity

• japanese-words-to-vectors - Word2vec (word to vectors) approach for Japanese language using Gensim and Mecab.
• chiVe - Japanese word embedding with Sudachi and NWJC
• elmo-japanese - elmo-japanese
• embedrank - Python Implementation of EmbedRank
• aovec - Easy aozorabunko Word2Vec Builder - 青空文庫全書籍のWord2Vecビルダー+構築済みモデル
• dependency-based-japanese-word-embeddings - This is a repository for the AI LAB article “係り受けに基づく日本語単語埋込 (Dependency-based Japanese Word Embeddings)” ( Article URL https://ai-lab.lapras.com/nlp/japanese-word-embedding/)
• jawikivec - Yet Another Japanese-Wikipedia Entity Vectors
• jawiki_word_vector_updater - 最新の日本語Wikipediaのダンプデータから，MeCabを用いてIPA辞書と最新のNeologd辞書の両方で形態素解析を実施し，その結果に基づいた word2vec，fastText，GloVeの単語分散表現を学習するためのスクリプト

Transformer based models

Models that use self-attention to understand context and perform advanced language tasks

• bert-japanese - BERT models for Japanese text.
• bert-japanese - BERT with SentencePiece for Japanese text.
• SudachiTra - Japanese tokenizer for Transformers
• japanese-dialog-transformers - Code for evaluating Japanese pretrained models provided by NTT Ltd.
• shiba - Pytorch implementation and pre-trained Japanese model for CANINE, the efficient character-level transformer.
• Dialog - A PyTorch Implementation of japanese chatbot using BERT and Transformer’s decoder
• language-pretraining - BERT and ELECTRA models of PyTorch implementations for Japanese text.
• medbertjp - Trials of pre-trained BERT models for the medical domain in Japanese.
• ILYS-aoba-chatbot - ILYS-aoba-chatbot
• t5-japanese - Codes to pre-train Japanese T5 models
• pytorch_bert_japanese - PytorchでBERTの日本語学習済みモデルを利用する
• Laboro-BERT-Japanese - Laboro BERT Japanese: Japanese BERT Pre-Trained With Web-Corpus
• RoBERTa-japanese - Japanese BERT Pretrained Model
• aMLP-japanese - aMLP Transformer Model for Japanese
• bert-japanese-aozora - Japanese BERT trained on Aozora Bunko and Wikipedia, pre-tokenized by MeCab with UniDic & SudachiPy
• sbert-ja - Code to train Sentence BERT Japanese model for Hugging Face Model Hub
• BERT-Japan-vaccination - Official fine-tuning code for “Emotion Analysis of Japanese Tweets and Comparison to Vaccinations in Japan”
• gpt2-japanese - Japanese GPT2 Generation Model
• text2text-japanese - gpt-2 based text2text conversion model
• gpt-ja - GPT-2 Japanese model for HuggingFace’s transformers
• albert-japanese - BERT with SentencePiece for Japanese text.
• ja_text_bert - 日本語WikipediaコーパスでBERTのPre-Trainedモデルを生成するためのリポジトリ
• DistilBERT-base-jp - A Japanese DistilBERT pretrained model, which was trained on Wikipedia.
• bert - This repository provides snippets to use RoBERTa pre-trained on Japanese corpus. Our dataset consists of Japanese Wikipedia and web-scrolled articles, 25GB in total. The released model is built based on that from HuggingFace.
• Laboro-DistilBERT-Japanese - Laboro DistilBERT Japanese
• luke - LUKE – Language Understanding with Knowledge-based Embeddings
• GPTSAN - General-purpose Swich transformer based Japanese language mode
• AcademicBART - We pretrained a BART-based Japanese masked language model on paper abstracts from the academic database CiNii Articles
• AcademicRoBERTa - We pretrained a RoBERTa-based Japanese masked language model on paper abstracts from the academic database CiNii Articles.
• LINE-DistilBERT-Japanese - DistilBERT model pre-trained on 131 GB of Japanese web text. The teacher model is BERT-base that built in-house at LINE.
• Japanese-Alpaca-LoRA - 日本語に翻訳したStanford Alpacaのデータセットを用いてLLaMAをファインチューニングし作成したLow-Rank AdapterのリンクとGenerateサンプルコード
• albert-japanese-tinysegmenter - Pretrained models, codes and guidances to pretrain official ALBERT(https://github.com/google-research/albert) on Japanese Wikipedia Resources
• japanese-llama-experiment - Japanese LLaMa experiment
• easylightchatassistant - EasyLightChatAssistant は軽量で検閲や規制のないローカル日本語モデルのLightChatAssistant を、KoboldCpp で簡単にお試しする環境です。

ChatGPT

Resources for using ChatGPT and APIs for Japanese dialogue and text generation

• VRChatGPT - ChatGPTを使ってVRChat上でお喋り出来るようにするプログラム。
• AITuberDegikkoMirii - AITuberの基礎となる部分を開発しています
• wanna - Shell command launcher with natural language
• ChatdollKit - ChatdollKit enables you to make your 3D model into a chatbot
• ChuanhuChatGPTJapanese - GUI for ChatGPT API For Japanese
• AISisterAIChan - ChatGPT3.5を搭載した伺かゴースト「AI妹アイちゃん」です。利用には別途ChatGPTのAPIキーが必要です。
• vrchatbot - VRChatにAI Botを作るためのリポジトリ
• gptuber-by-langchain - GPTがYouTuberをやります
• openai-chatfriend - A chatbox application built using Nuxt 3 powered by Open AI Text completion endpoint. You can select different personality of your AI friend. The default will respond in Japanese. You can use this app to practice your Nihongo skills!
• chrome-ext-translate-to-hiragana-with-chatgpt - This Chrome extension can translate selected Japanese text to Hiragana by using ChatGPT.
• azure-search-openai-demo - このサンプルでは、Retrieval Augmented Generation パターンを使用して、独自のデータに対してChatGPT のような体験を作成するためのいくつかのアプローチを示しています。
• chatvrm - ChatVRMはブラウザで簡単に3Dキャラクターと会話ができるデモアプリケーションです。
• sftly-replace - A Chrome extention to replace the selected text softly
• summarize_arxv - Summarize arXiv paper with figures
• aiavatarkit - Building AI-based conversational avatars lightning fast
• jp-azureopenai-samples - Azure OpenAIを活用したアプリケーション実装のリファレンスを目的として、アプリのサンプル（リファレンスアーキテクチャ、サンプルコードとデプロイ手順）を無償提供しています。
• character_chat - OpenAIのAPIを利用して、設定したキャラクターと日本語で会話するチャットスクリプトです。
• chatgpt-slackbot - OpenAIのChatGPT APIをSlack上で利用するためのSlackbotスクリプト (日本語での利用が前提)
• chatgpt-prompt-sample-japanese - ChatGPT の Prompt のサンプルです。
• kanji-flashcard-app-gpt4 - A Japanese Kanji Flashcard App built using Python and Langchain, enhanced with the intelligence of GPT-4.
• IgakuQA - Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing Examinations
• japagen - 日本語タスクにおけるLLMを用いた疑似学習データ生成の検討
• generativeai-prompt-sample-japanese - ChatGPTやCopilotなど各種生成AI用の「日本語]の Prompt のサンプル

Dictionary and IME

Resources for Japanese dictionaries and input method editors (IME)

• mecab-ipadic-neologd - Neologism dictionary based on the language resources on the Web for mecab-ipadic
• tdmelodic - A Japanese accent dictionary generator
• jamdict - Python 3 library for manipulating Jim Breen’s JMdict, KanjiDic2, JMnedict and kanji-radical mappings
• unidic-py - Unidic packaged for installation via pip.
• Japanese-Company-Lexicon - Japanese Company Lexicon (JCLdic)
• manbyo-sudachi - Sudachi向け万病辞書
• jawiki-kana-kanji-dict - Generate SKK/MeCab dictionary from Wikipedia(Japanese edition)
• JIWC-Dictionary - dictionary to find emotion related to text
• JumanDIC - This repository contains source dictionary files to build dictionaries for JUMAN and Juman++.
• ipadic-py - IPAdic packaged for easy use from Python.
• unidic-lite - A small version of UniDic for easy pip installs.
• emoji-ime-dictionary - 日本語で絵文字入力をするための IME 追加辞書 orange_book Google 日本語入力などで日本語から絵文字への変換を可能にする IME 拡張辞書
• google-ime-dictionary - 日英変換・英語略語展開のための IME 追加辞書 orange_book 日本語から英語への和英変換や英語略語の展開を Google 日本語入力や ATOK などで可能にする IME 拡張辞書
• dic-nico-intersection-pixiv - ニコニコ大百科とピクシブ百科事典の共通部分のIME辞書
• google-ime-user-dictionary-ja-en - GoogleIME用カタカナ語辞書プロジェクトのアーカイブです。Project archive of Google IME user dictionary from Katakana word ( Japanese loanword ) to English.
• emoticon - Google日本語入力の顔文字辞書∩(,Ò‿Ó,)∩
• mecab-mozcdic - open source mozc dictionaryをMeCab辞書のフォーマットに変換したものです。
• denonbu-ime-dic - 電音IME: Microsoft IMEなどで利用することを想定した「電音部」関連用語の辞書
• nijisanji-ime-dic - Microsoft IMEなどで利用することを想定した「にじさんじ」関連用語の用語辞書です。
• pokemon-ime-dic - Microsoft IMEなどで利用することを想定した、現状判明している全てのポケモンの名前を網羅した用語辞書です。
• EJDict - English-Japanese Dictionary data (Public Domain) EJDict-hand
• Ayashiy-Nipongo-Dic - 贵樣ばこゐ辞畫を使て正レい日本语を使ラことが出來ゑ。
• genshin-dict - Windows/macOSで使える原神の単語辞書です
• jmdict-simplified - JMdict and JMnedict in JSON format
• mozcdict-ext - Convert external words into Mozc system dictionary
• mh-dict-jp - MonsterHunterのユーザー辞書を作りたい…
• mecab-unidic-neologd - Neologism dictionary based on the language resources on the Web for mecab-unidic
• hololive-dictionary - ホロライブ（ホロライブプロダクション）に関する辞書ファイルです。./dictionary フォルダ内のテキストファイルを使って、IMEに単語を追加できます。詳細はREADME.mdをご覧ください。
• jmdict-yomitan - JMdict, JMnedict, KANJIDIC for Yomitan/Yomichan.
• yomichan-jlpt-vocab - JLPT level tags for words in Yomichan
• Jitendex - A free and openly licensed Japanese-to-English dictionary compatible with multiple dictionary clients
• jiten - japanese android/cli/web dictionary based on jmdict/kanjidic — 日本語　辞典　和英辞典　漢英字典　和独辞典　和蘭辞典
• pixiv-yomitan - Pixiv Encyclopedia Dictionary for Yomitan
• uchinaaguchi_dict - うちなーぐち辞典（沖縄語辞典）
• yomitan-dictionaries - Japanese and Chinese dictionaries for Yomitan.
• mouse_over_dictionary - マウスオーバーした単語を自動で読み取る汎用辞書ツール
• jisyo - かな漢字変換エンジン SKKのための新しい辞書形式
• skk-jisyo.emoji-ja - 日本語の読みから Emoji に変換するための SKK 辞書 😂
• anthy - Anthy is a kana-kanji conversion engine for Japanese. It converts roma-ji to kana, and the kana text to a mixed kana and kanji.
• aws_dic_for_google_ime - AWSサービス名のGoogle日本語入力向けの辞書
• cl-skkserv - Common LispによるSKK辞書サーバーとその拡張
• anthy - Anthy maintenance
• anthy-unicode - Anthy Unicode - Another Anthy
• azooKey - azooKey is an open-source Japanese keyboard for iPhone and iPad, written in Swift and powered by its own kana-kanji conversion engine. It provides live conversion, flexible key layouts, and a clean SwiftUI interface for a smooth typing experience.
• azooKey-Desktop - azooKey-Desktop is an open-source Japanese input method for macOS, written in Swift and powered by the Zenzai neural kana-kanji converter. It provides live conversion, optional LLM-based “Magic Conversions”, and Tuner-backed personalization for a smooth, desktop typing experience.
• fcitx5-hazkey - Japanese input method for fcitx5, powered by azooKey engine
• mozcdic-ut-place-names - Mozc UT Place Name Dictionary is a dictionary converted from the Japan Post’s ZIP code data for Mozc.
• AzooKeyKanaKanjiConverter - Kana-Kanji Conversion Module written in Swift, supporting Neural Kana-Kanji Conversion and other cool features.
• libkkc - Japanese Kana Kanji conversion input method library
• libskk - Japanese SKK input method library
• cjkvi-dict - 漢字データベースの辞書関連データ
• wlsp-classical - 古典日本語の分類語彙表データ
• kanji-dict - 漢字の書き順(筆順)・読み方・画数・部首・用例・成り立ちを調べるための漢字辞書です。Unicode 15.1 のすべての漢字 98,682字を収録しています。
• Kaomoji_proj - (๑ ᴖ ᴑ ᴖ ๑)みょんかおもじ（旧Kaomoji_proj）はMicrosoft社の入力ソフト、Microsoft IME向けの顔文字の辞書を作成するプロジェクトです。
• kotlin-kana-kanji-converter - Kotlin かな漢字変換プログラム
• alfred-japanese-dictionary - Japanese-English Dictionary using jisho.org with audio, csv export of entries, and preview of dictionary sites.
• ichiran - Linguistic tools for texts in Japanese language
• mikan - A Japanese input method.
• colloquial-kansai-dictionary - A quick reference for the material taught in Colloquial Kansai Japanese.
• jisho-open - Web frontend for the JMdict Japanese-English dictionary project, with study list support!
• macskk - Yet Another macOS SKK Input Method
• nandoku - 難読漢字を学年別にまとめた辞書です。
• japanese_android_ime - A FOSS Japanese IME for Android
• anthywl - Japanese input method for Sway using libanthy
• sekka - Yet another Japanese Input Method inspired by SKK.
• sumibi - Japanese input method powered by ChatGPT API
• jinmei-dict - 辞書データから人名だけを抜き出し、読み仮名（カタカナ）をキーとして、候補となる書き文字をリストで保持するようなJSON形式に整形しています。
• japanesearabic - JapaneseArabic Dictionary (日本語・アラビア語辞書) قاموس اللغة اليابانية والعربية (Yomitan)
• o-dic - 沖縄辞書
• skk-emoji-jisyo - SKK 絵文字辞書
• mozcdic-ut-personal-names - A personal name dictionary for Mozc.
• mozcdic-ut-sudachidict - A dictionary converted from SudachiDict for Mozc.
• nihongo - japanese language data and dictionary
• kagome-dict - Dictionary Library for Kagome v2
• canna - Canna Japanese input system
• kansai-accent-dictionary - 京阪式アクセント（関西弁）辞書 - 4,615語を収録した日本語方言アクセント辞書
• jitendex - A free, offline, and openly licensed Japanese-to-English dictionary. Updates monthly!
• karukan - Japanese Input Method System for Linux, Neural Kana-Kanji Conversion Engine + fcitx5 IME
• shitto-mania-dic - 嫉妬辞書（Shitto-Mania / Jealousy Dictionary）
• dvorakjp-roman-table - azooKey, Google 日本語入力用 DvorakJP ローマ字テーブル / DvorakJP Roman Table for azooKey, Google Japanese Input
• jmdict-fst - Fast JMdict lookup engine with FST-based exact/prefix/fuzzy/gloss search, deinflection, Rust core, and Swift/Kotlin/Flutter bindings.
• mzimeja - MZ-IME Japanese Input for Windows
• japanesekeyboard - スミレ - 完全オフラインの日本語キーボードアプリ
• Sumibi - Japanese/Chinese input method powered by ChatGPT API
• rakukan - ローカルLLMを利用した、Windows 向け日本語 IMEgit
• JMdictSQLite - SQLite database for JMdict and Kanjidic, a Japanese-English dictionary. Automatic daily updates.
• Meltype - 半角/全角 キーを押さなくても、日本語と英語を打ち分けられるようにする Windows 常駐ツールです。

Corpus

Part-of-speech tagging / Named entity recognition

Corpora annotated with part-of-speech tags and named entities

• ner-wikipedia-dataset - Wikipediaを用いた日本語の固有表現抽出データセット
• IOB2Corpus - Japanese IOB2 tagged corpus for Named Entity Recognition.
• TwitterCorpus - 首都大日本語 Twitter コーパス
• UD_Japanese-PUD - Parallel Universal Dependencies.
• UD_Japanese-GSD - Japanese data from the Google UDT 2.0.
• KWDLC - Kyoto University Web Document Leads Corpus
• AnnotatedFKCCorpus - Annotated Fuman Kaitori Center Corpus
• UD_Japanese-GSDLUW - Long-unit-word version of UD_Japanese-GSD
• ud_japanese-bccwj - This Universal Dependencies (UD) Japanese treebank is based on the definition of UD Japanese convention described in the UD documentation.

Parallel corpus

Bilingual corpora containing aligned sentences for translation tasks

• small_parallel_enja - 50k English-Japanese Parallel Corpus for Machine Translation Benchmark.
• Web-Crawled-Corpus-for-Japanese-Chinese-NMT - A Web Crawled Corpus for Japanese-Chinese NMT
• CourseraParallelCorpusMining - Coursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures Translation
• JESC - A large parallel corpus of English and Japanese
• AMI-Meeting-Parallel-Corpus - AMI Meeting Parallel Corpus
• giant_ja-en_parallel_corpus - This directory includes a giant Japanese-English subtitle corpus. The raw data comes from the Stanford’s JESC project.
• jesc_small - Small Japanese-English Subtitle Corpus
• graded-enja-corpus - 禁止用語や単語レベルを考慮した日英対訳コーパスです。
• cjk-compsci-terms - CJK computer science terms comparison / 中日韓電腦科學術語對照 / 日中韓のコンピュータ科学の用語対照 / 한·중·일 전산학 용어 대조
• Laboro-ParaCorpus - Scripts for creating a Japanese-English parallel corpus and training NMT models
• google-vs-deepl-je - google-vs-deepl-je
• matcha - 訪日観光客向けメディアMATCHAの記事から、日本語のテキスト平易化のためのデータセットを構築しました。
• en-ja-el - EnJaEL: En-Ja Parallel Entity Linking Dataset (Version 1.0)

Dialog corpus

Collections of conversation data for training dialogue systems

• JMRD - Japanese Movie Recommendation Dialogue dataset
• open2ch-dialogue-corpus - おーぷん2ちゃんねるをクロールして作成した対話コーパス
• BSD - The Business Scene Dialogue corpus
• asdc - Accommodation Search Dialog Corpus (宿泊施設探索対話コーパス)
• japanese-corpus - 日本語の対話データ for seq2seq etc
• BPersona-chat - This repository contains the Japanese–English bilingual chat corpus BPersona-chat published in the paper Chat Translation Error Detection for Assisting Cross-lingual Communications at AACL-IJCNLP 2022’s Workshop Eval4NLP 2022.
• japanese-daily-dialogue - Japanese Daily Dialogue, or 日本語日常対話コーパス in Japanese, is a high-quality multi-turn dialogue dataset containing daily conversations on five topics: dailylife, school, travel, health, and entertainment.
• llm-japanese-dataset - LLM構築用の日本語チャットデータセット
• kokorochat - ロールプレイで収集した日本語のカウンセリング対話データセット
• JMultiWOZ-TC - マルチターン対話でのエージェントのfunction calling評価
• HOTATE - 本音・建前付き日本語対話データセット
• ETCDataset - Emotion Transcription in Conversation Dataset は，対話中の各発話に対して話者自身が記述した心情文を含む，約1,000 件の対話からなる日本語対話データセットです．

Others

Corpora for tasks such as question answering or entailment recognition

• jrte-corpus - Japanese Realistic Textual Entailment Corpus (NLP 2020, LREC 2020)
• kanji-data - A JSON kanji dataset with updated JLPT levels and WaniKani information
• JapaneseWordSimilarityDataset - Japanese Word Similarity Dataset
• simple-jppdb - A paraphrase database for Japanese text simplification
• chABSA-dataset - chakki’s Aspect-Based Sentiment Analysis dataset
• JaQuAD - JaQuAD: Japanese Question Answering Dataset for Machine Reading Comprehension (2022, Skelter Labs)
• JaNLI - Japanese Adversarial Natural Language Inference Dataset
• ebe-dataset - Evidence-based Explanation Dataset (AACL-IJCNLP 2020)
• emoji-ja - UNICODE絵文字の日本語読み/キーワード/分類辞書
• nayose-wikipedia-ja - Wikipediaから作成した日本語名寄せデータセット
• ja.text8 - Japanese text8 corpus for word embedding.
• ThreeLineSummaryDataset - 3行要約データセット
• japanese - This repo contains a list of the 44,998 most common Japanese words in order of frequency, as determined by the University of Leeds Corpus.
• kanji-frequency - Kanji usage frequency data collected from various sources
• TEDxJP-10K - TEDxJP-10K ASR Evaluation Dataset
• CoARiJ - Corpus of Annual Reports in Japan
• technological-book-corpus-ja - 日本語で書かれた技術書を収集した生コーパス/ツール
• ita-corpus-chuwa - Chunked word annotation for ITA corpus
• wikipedia-utils - Utility scripts for preprocessing Wikipedia texts for NLP
• inappropriate-words-ja - 日本語における不適切表現を収集します。自然言語処理の時のデータクリーニング用等に使えると思います。
• house-of-councillors - 参議院の公式ウェブサイトから会派、議員、議案、質問主意書のデータを整理しました。
• house-of-representatives - 国会議案データベース：衆議院
• STAIR-captions - STAIR captions: large-scale Japanese image caption dataset
• Winograd-Schema-Challenge-Ja - Japanese Translation of Winograd Schema Challenge
• speechBSD - An extension of the BSD corpus with audio and speaker attribute information
• ita-corpus - ITAコーパスの文章リスト
• rohan4600 - モーラバランス型日本語コーパス
• anlp-jp-history - 言語処理学会年次大会講演の全リスト・機械可読版など
• keigo_transfer_task - 敬語変換タスクにおける評価用データセット
• loanwords_gairaigo - English loanwords in Japanese
• jawikicorpus - Japanese-Wikipedia Wikification Corpus
• GeneralPolicySpeechOfPrimeMinisterOfJapan - This is the corpus of Japanese Text that general policy speech of prime minister of Japan
• wrime - WRIME: 主観と客観の感情分析データセット
• jtubespeech - JTubeSpeech: Corpus of Japanese speech collected from YouTube
• WikipediaWordFrequencyList - 日本語Wikipediaで使用される頻出単語のリスト
• kokkosho_data - 車両不具合情報に関するデータセット
• pdmocrdataset-part1 - デジタル化資料OCRテキスト化事業において作成されたOCR学習用データセット
• huriganacorpus-ndlbib - 全国書誌データから作成した振り仮名のデータセット
• jvs_hiho - JVS (Japanese versatile speech) コーパスの自作のラベル
• hirakanadic - Allows Sudachi to normalize from hiragana to katakana from any compound word list
• animedb - 約100年に渡るアニメ作品リストデータベース
• security_words - サイバーセキュリティに関連する公的な組織の日英対応
• Data-on-Japanese-Diet-Members - 日本の国会議員のデータ
• honkoku-data - 歴史資料の市民参加型翻刻プラットフォーム「みんなで翻刻」のテキストデータ置き場です。 / Transcription texts created on Minna de Honkoku (https://honkoku.org), a crowdsourced transcription platform for historical Japanese documents.
• engineer-vocabulary-list - Engineer Vocabulary List in Japanese/English
• JSICK - Japanese Sentences Involving Compositional Knowledge (JSICK) Dataset/JSICK-stress Test Set
• phishurl-list - Phishing URL dataset from JPCERT/CC
• jcms - A Japanese Corpus of Many Specialized Domains (JCMS)
• aozorabunko_text - text-only archives of www.aozora.gr.jp
• topokanji - Topologically ordered lists of kanji for effective learning
• isbn4groups - ISBN-13における日本語での出版物 (978-4-XXXXXXXXX) に関するデータ等
• NMeCab - NMeCab: About Japanese morphological analyzer on .NET
• ndlngramdata - デジタル化資料から作成したOCRテキストデータのngram頻度統計情報のデータセット
• ndlngramviewer_v2 - 2023年1月にリニューアルしたNDL Ngram Viewerのソースコード等一式
• data_set - 法律・判例関係のデータセット
• huggingface-datasets_wrime - WRIME for huggingface datasets
• ndl-minhon-ocrdataset - NDL古典籍OCR学習用データセット（みんなで翻刻加工データ）
• PAX_SAPIENTICA - GIS & Archaeological Simulator. 2023 in development.
• j-liwc2015 - Japanese version of LIWC2015
• huggingface-datasets_livedoor-news-corpus - Japanese Livedoor news corpus for huggingface datasets
• huggingface-datasets_JGLUE - JGLUE: Japanese General Language Understanding Evaluation for huggingface datasets
• commonsense-moral-ja - JCommonsenseMorality is a dataset created through crowdsourcing that reflects the commonsense morality of Japanese annotators.
• comet-atomic-ja - COMET-ATOMIC ja
• dcsg-ja - Dialogue Commonsense Graph in Japanese
• japanese-toxic-dataset - “Proposal and Evaluation of Japanese Toxicity Schema” provides a schema and dataset for toxicity in the Japanese language.
• camera - CAMERA (CyberAgent Multimodal Evaluation for Ad Text GeneRAtion) is the Japanese ad text generation dataset.
• Japanese-Fakenews-Dataset - 日本語フェイクニュースデータセット
• copa-japanese - COPA Dataset in Japanese
• WLSP-familiarity - Word Familiarity Rate for ‘Word List by Semantic Principles (WLSP)’
• ProSub - A cross-linguistic study of pronoun substitutes and address terms
• ramendb - なんとかデータベース( https://supleks.jp/ )からのスクレイピングツールと収集データ
• huggingface-datasets_CAMERA - CAMERA (CyberAgent Multimodal Evaluation for Ad Text GeneRAtion) for huggingface datasets
• FactCheckSentenceNLI-FCSNLI- - FactCheckSentenceNLIデータセット
• databricks-dolly-15k-ja - databricks/dolly-v2-12b の学習データに使用されたdatabricks-dolly-15k.jsonl を日本語に翻訳したデータセットになります。
• EaST-MELD - EaST-MELD is an English-Japanese dataset for emotion-aware speech translation based on MELD.
• meconaudio - Mecon Audio(Medical Conference Audio)は厚生労働省主催の先進医療会議の議事録の読み上げデータセットです。
• japanese-addresses - 全国の町丁目レベル（277,191件）の住所データのオープンデータ
• aozorasearch - The full-text search system for Aozora Bunko by Groonga. 青空文庫全文検索ライブラリ兼Webアプリ。
• llm-jp-corpus - This repository contains scripts to reproduce the LLM-jp corpus.
• alpaca_ja - alpacaデータセットを日本語化したものです
• instruction_ja - Japanese instruction data (日本語指示データ)
• japanese-family-names - Top 5000 Japanese family names, with readings, ordered by frequency.
• kanji-data-media - Japanese language data on kanji, radicals, media files, fonts and related resources from Kanji alive
• reazonspeech - Construct large-scale Japanese audio corpus at home
• huriganacorpus-aozora - 青空文庫及びサピエの点字データから作成した振り仮名のデータセット
• koniwa - An open collection of annotated voices in Japanese language
• JMMLU - 日本語マルチタスク言語理解ベンチマーク Japanese Massive Multitask Language Understanding Benchmark
• hurigana-speech-corpus-aozora - 青空文庫振り仮名注釈付き音声コーパスのデータセット
• jqara - JQaRA: Japanese Question Answering with Retrieval Augmentation - 検索拡張(RAG)評価のための日本語Q&Aデータセット
• jemhopqa - JEMHopQA (Japanese Explainable Multi-hop Question Answering) is a Japanese multi-hop QA dataset that can evaluate internal reasoning.
• jacred - Repository for Japanese Document-level Relation Extraction Dataset (plan to be released in March).
• jades - JADES is a dataset for text simplification in Japanese, described in “JADES: New Text Simplification Dataset in Japanese Targeted at Non-Native Speakers” (the paper will be available soon).
• do-not-answer-ja - 2023年8月にメルボルン大学から公開された安全性評価データセット『Do-Not-Answer』を日本語LLMの評価においても使用できるように日本語に自動翻訳し、さらに日本文化も考慮して修正したデータセット。
• oasst1-89k-ja - OpenAssistant のオープンソースデータ OASST1 を日本語に翻訳したデータセットになります。
• jacwir - JaCWIR: Japanese Casual Web IR - 日本語情報検索評価のための小規模でカジュアルなWebタイトルと概要のデータセット
• japanese-technical-dict - 日本語学習者のための科学技術業界でよく使われる片仮名と元の単語対照表
• j-unimorph - Dataset of UniMorph in Japanese
• GazeVQA - Dataset for the LREC-COLING 2024 paper “A Gaze-grounded Visual Question Answering Dataset for Clarifying Ambiguous Japanese Questions”
• J-CRe3 - Code for J-CRe3 experiments (Ueda et al., LREC-COLING, 2024)
• jmed-llm - JMED-LLM: Japanese Medical Evaluation Dataset for Large Language Models
• lawtext - Plain text format for Japanese law
• pdmocrdataset-part2 - OCR処理プログラム研究開発事業において作成されたOCR学習用データセット
• japanesetopicwsd - 話題に基づく語義曖昧性解消評価セット
• temporalNLI_dataset - Jamp: Controlled Japanese Temporal Inference Dataset for Evaluating Generalization Capacity of Language Models
• JSeM - Japanese semantic test suite (FraCaS counterpart and extensions)
• niilc-qa - NIILC QA data
• chain-of-thought-ja-dataset - Dataset of paper “Verification of Chain-of-Thought Prompting in Japanese”
• WikipediaAnnotatedCorpus - This is a Japanese text corpus that consists of Wikipedia articles with various linguistic annotations.
• elaws-history - e-Gov 法令検索で配布されている「全ての法令データ」を定期的にダウンロードし、アーカイブしています
• Japanese-RP-Bench - Japanese-RP-BenchはLLMの日本語ロールプレイ能力を測定するためのベンチマークです。
• hdic - HDIC : Integrated Database of Hanzi Dictionaries in Early Japan
• awesome-japan-opendata - Awesome Japan Open Data - 日本のオープンデータ情報一覧・まとめ
• kanji-data - 常用漢字表他、漢字に関するデータ
• openchj-genji - 「源氏物語」形態論情報データ
• AdParaphrase - This repository contains data for our paper “AdParaphrase: Paraphrase Dataset for Analyzing Linguistic Features toward Generating Attractive Ad Texts”.
• Jamp_sp - アスペクトを考慮した日本語時間推論データセットの構築（Jamp_sp: Controlled Japanese Temporal Inference Dataset Considering Aspect）
• jnli-neg - 否定理解能力を評価するための日本語言語推論データセット JNLI-Neg の公開用リポジトリです。
• swallow-corpus - This repository provides Python implementation for building Swallow Corpus Version 1, a large Japanese web corpus (Okazaki et al., 2024), from Common Crawl archives.
• jalecon - A Dataset of Japanese Lexical Complexity for Non-Native Readers
• multils-japanese - MultiLS-Japanese Lexical Complexity Prediction and Lexical Simplification Dataset for Japanese: annotator profiles, unaggregated annotation, and annotatation guidelines.
• nwjc - NINJAL Web Japanese Corpus
• open-mantra-dataset - Dataset introduced in the paper “Towards Fully Automated Manga Translation” presented in AAAI21
• public-annotations - Various annotations of Manga109 dataset
• gimei - random Japanese name and address generator
• safety-boundary-test - 日本語言語モデルの安全性の振る舞いを評価するテストセット
• j-ono-data - A simple, open-source collection of Japanese onomatopoeic and mimetic sound words in JSON format. With manga samples.
• kanji - List of japanese kanji radicals to learn
• jethics - 日本語道徳理解度評価用データセットJETHICSの概説ページ (to be update)
• waon - WAON: Large-Scale and High-Quality Japanese Image-Text Dataset for Vision-Language Models
• kuci - Kyoto University Commonsense Inference dataset (KUCI)
• japanese-address-testdata - 解析が難しい日本の住所のテストデータセット
• jlpt-word-list - Japanese word list from JLPT vocabulary
• hiragana_mojigazo - 文字画像データセット(平仮名73文字版)
• lawqa_jp - 日本の法令に関する多肢選択式QAデータセット
• yjcaptions - YJ Captions 26k Dataset
• ja-vg-vqa - Japanese Visual Genome VQA dataset
• lawhub - Repository to track Japanese Law in text format
• jconj - A table-based Japanese word conjugator
• extract_jawp_names - Extracts personal names in Wikipedia Japanese.
• cejc_yomichan_freq_dict - Frequency dictionary for yomichan based on the Corpus of Everyday Japanese Conversation dataset
• wikidict-ja - Wikipedia Bilingual Reference Data (Japanese)
• ajimee-bench - AJIMEE-Bench (Advanced Japanese IME Evaluation Benchmark)
• j-spaw - J-SpAW: Japanese speech corpus for speaker verification and anti-spoofing
• camera3 - CAMERA3: An Evaluation Dataset for Controllable Ad Text Generation in Japanese
• jgpqa - Japanese translation of the GPQA dataset
• tanaka-corpus-plus - Tanaka Corpus のノイズを除去しています。
• emotioncorpusjapanesetokushimaa2lab - Japanese emotion corpus Tokushima Univ. A-2 Lab.
• osworld-jp - 言語を考慮した評価のための、日本語版コンピュータユースベンチマーク
• quasi_japanese_reviews - Quasi Japanese Reviews (擬似レビューデータ)
• psychiatry-clinical-notes - 精神科初診カルテ作成アンケート データセット
• merged-town-names - 市町村合併などにより消滅した旧地名と新地名の対応表
• japanesetextemoticondata - Japanese text-emoticon data.
• mishearing-corpus - 聞き間違えコーパス︱CSV＋Table Schema で約 1 万件を管理し、VS Code＋pre-commit＋Frictionless＋GitHub Actions で自動検証を行う日本語データセット
• kotowaza - Structured JSON dataset of Japanese proverbs (kotowaza) with meanings in Indonesian & English, examples, JLPT levels, and tags.
• selective-rag-kasensabo - 建設の技術基準に関する質問の専門性粒度（細かい/粗い）を96%正確に自動判定し、最適なRAGシステム（ColBERT/Naive）を選択する実用的なAgentic RAGシステムのMVPです。2025年11月に公開された河川砂防ダムの技術基準を対象に４つのRAGシステムを構築し、専門性の粒度が異なる200問の質問に対して、精度と速度を比較した。
• jmle2026-bench - LLM benchmark on the 120th Japanese Medical Licensing Examination (Feb 7-8, 2026)
• JSTS-Neg - 否定理解能力を評価するための日本語意味的類似度計算データセット JSTS-Neg の公開用リポジトリです。 JSTS-Neg は、JGLUE に含まれる言語推論データセット JSTS を拡張して作成しました。
• business-slide-questions - このリポジトリでは、ビジネス資料（スライド）を対象とした Visual Question Answering (VQA) ベンチマーク「BusinessSlideVQA」を提供しています。
• WLSP-antonym - Antonym relations for ‘Word List by Semantic Principles (WLSP)’
• YouCook2-JP - Japanese translation of the YouCook2 dataset.
• E2U - つたわる化に関するデータ
• annotation-2025 - このリポジトリは，テキストの「解釈」を人手とLLM出力で比較できるデータを公開するためのものです．
• jhpt - 歴史的日本語資料の原文テキストと，現代語訳（参照訳）テキストをセグメント単位で対応付けた対訳データセットです．詳細は論文を参照ください．
• JBE-QA - Japanese Bar Exam QA
• JMedWiC - マスク言語モデルを用いて擬似的な同義・非同義ペアを自動抽出し，人手による同義性アノテーションを通じてラベルを決定することで，日本語の医療分野における語義同一性判定データセットを構築しました．
• Doppelganger-JC - This is a dataset benchmarking the misuse of cross-lingual homographs between Chinese and Japanese in LLMs.
• modelvista-3lang - ソフトウェア図理解のためのVLM評価ベンチマーク（日本語・英語・韓国語対応）
• japanese-hr-niah - 日本語人事労務ドメインにおけるロングコンテキストLLMの性能評価ベンチマーク
• nijl-manyoshutei - 本リポジトリでは、関西大学所蔵廣瀬本万葉集のTEI/XMLデータ等をCC-BYライセンスのもとで公開しています。
• kamuskita - マレー語勉強会で作っているオープンなマレー語・日本語辞典『みんなのマレー語辞典』
• japanese-llm-benchmark - A benchmark tool for evaluating Japanese language capabilities of various LLMs.
• EDINET-Bench - ICLR 2026 Evaluating the performance of LLMs on Japanese challenging financial tasks.
• LCTG-Bench - LCTG Bench: LLM Controlled Text Generation Benchmark
• Kokoro-Speech-Dataset - A public domain single speaker Japanese speech dataset
• LookVQA - A Gaze-grounded Visual Question Answering Dataset for Clarifying Ambiguous Japanese Questions (LREC-COLING 2024)
• JTruthfulQA - JTruthfulQA is a Japanese version of TruthfulQA (Lin+, 2022). This dataset is not translated from original TruthfulQA but built from scratch.
• japanese-dataset-for-automated-fact-checking - Japanese Dataset for Automated Fact-Checking: JAD-AFC
• llm-jp-longbench - 日本語版longbench作成のため、jemhopデータセットを活用。
• doctrine-corpus - Bilingual (Japanese + English) judgment-eliciting Q&A corpus (851 examples) encoding documented research-program decisions, with per-example metadata (CC0, Hugging Face mirror available)
• medLLM_QA_benchmark - Trilingual (English, Japanese, Chinese) QA benchmark for medical LLM
• kaomoji-data - 約1,700種類の日本語顔文字を全件、かな読み（擬態語・キーワード）・タグ・カテゴリでアノテーションした構造化JSONデータセット。多カテゴリ対応・カテゴリ別ファイル分割、MITライセンス。
• jvs_nonpara_kana - Katakana annotation of JVS nonpara corpus for G2P evaluation
• jlpt-kanji-dictionary - Structured Japanese Kanji and Vocabulary JSON datasets organized by JLPT level with English and Russian translations — ready for use in language learning apps, NLP, and kanji study tools.
• pfgen-bench - Preferred Generation Benchmark
• j-tau-bench - J-tau: A Japanese tau-bench for Benchmarking Tool-Agent-User Interaction in Real-World Domains
• bbh-ja - Japanese Translation of BIG-Bench-Hard (https://github.com/suzgunmirac/BIG-Bench-Hard/)
• jfbench - JFBench: Japanese instruction Following Benchmark
• aica-corpus - AIキャラクター・フィラー・笑い声・感情表現に特化した日本語TTSコーパス（CC0）
• adlib - ADLIB: Japanese ASR benchmark framework with language-aware evaluation
• jmeeteval - J-MeetEval: 日本語会議タスクにおける指示追従性評価ベンチマーク