JETHICS
Models judge ethical scenarios in Japanese. The suite includes the Japanese commonsense morality dataset used by llm-jp-eval.
The authors provide evaluation samples and a separate download of the full dataset.
- Made by
- Masashi Takeshita and Rafal Rzepka
- Released
- June 2025
- Kind
- Reasoning in many languages
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Philosophy, ethics and religion
- Language
- Japanese
- Interaction
- Single turn
- Items from
- Another benchmark
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-4o Nov 2024 #161 on the capability index
- Last model added
- November 2024
- Scores refreshed
- November 20th, 2026
GPT-4o leads with 71.3% GPT-4o leads with 71.3%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
- 1 GPT-4o Nov 2024, 8-shot 71.3%
- 2 GPT-4o mini 8-shot 50.8%
- 3 llm-jp-3-13b-instruct 8-shot 49.7%
- 4 Llama-3-ELYZA-JP-8B 8-shot 47.9%
- 5 Meta-Llama-3-8B-Instruct 8-shot 36.7%
- 6 llm-jp-3-3.7b-instruct 8-shot 32.6%
Source: The JETHICS paper
- 1 GPT-4o Nov 2024, 8-shot 71.3%
- 2 GPT-4o mini 8-shot 50.8%
Source: The JETHICS paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o Nov 2024 | 71.3% | 8-shot | Nov 2024 | Official leaderboard |
| 2 | GPT-4o mini | 50.8% | 8-shot | Jul 2024 | Official leaderboard |
| 3 | llm-jp-3-13b-instruct | 49.7% | 8-shot | Official leaderboard | |
| 4 | Llama-3-ELYZA-JP-8B | 47.9% | 8-shot | Official leaderboard | |
| 5 | Meta-Llama-3-8B-Instruct | 36.7% | 8-shot | Official leaderboard | |
| 6 | llm-jp-3-3.7b-instruct | 32.6% | 8-shot | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o Nov 2024 | 71.3% | 8-shot | Nov 2024 | Official leaderboard |
| 2 | GPT-4o mini | 50.8% | 8-shot | Jul 2024 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official JETHICS leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in November 2024, for GPT-4o. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (6 entries).
- Pass Models scored 2 models we track, plus 4 other entries on its leaderboard.
- Warning Newest score The newest came 2 years ago, in November 2024, for GPT-4o.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-4o (#161).
- Pass Headroom The best is 71.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 2 rows for models we track and 4 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.