Kaggle Global-MMLU-Lite - Japanese: leaderboard
Accuracy on the Japanese questions (400 on this board) of Cohere Labs' Global-MMLU-Lite leaderboard on Kaggle, an MMLU-derived multilingual multiple-choice set with culturally sensitive and culturally agnostic questions. Shown as 0-100, higher is better; four answer options, so guessing scores about 25.
Metric: Accuracy (%). Source: www.kaggle.com. Saturation forecast: Estimated already saturated. 44 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 96.25 | #9 |
| 2 | Claude Opus 5 | 96 | #16 |
| 3 | Gemini 3.5 Flash | 95.75 | #66 |
| 4 | GPT-6 Sol | 95.75 | #19 |
| 5 | GPT-6 Astra | 95.5 | #6 |
| 6 | Claude Opus 5.5 | 95 | #3 |
| 7 | GPT-5.6 Sol | 94.75 | #22 |
| 8 | Claude Opus 4.8 | 94.75 | #39 |
| 9 | Gemini 3.8 Flash | 94.75 | #30 |
| 10 | Gemini 3.6 Flash | 94.75 | #57 |
| 11 | Grok 4.5 | 94.75 | #54 |
| 12 | Gemini 3.7 Flash | 94.5 | #24 |
| 13 | Gemini 3 Pro (Preview) | 94 | #71 |
| 14 | Claude Opus 4.1 | 94 | #146 |
| 15 | Gemini 2.5 Pro | 93.75 | #159 |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=kaggle-global-mmlu-lite-japanese · How It Works · Data refreshed daily, snapshot 2026-11-05.