JMMMU-Pro - World History: leaderboard
The "World History" subtask of JMMMU-Pro. Japanese multimodal understanding benchmark extending JMMMU with image-based multidisciplinary questions.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 96.67 | #80 |
| 2 | GPT-5.2 | 90.67 | #110 |
| 3 | Qwen 2.5 VL 7B | 75.33 | #643 |
| 4 | aya-vision-8B | 28.67 | #1107 |
| 5 | Pangea-7B | 19.33 | #1224 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=jmmmu-pro-world-history · How It Works · Data refreshed daily, snapshot 2026-10-20.