MEGA-Bench Task - Multilingual Ruozhiba Explanation Japanese — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 76.4 |
| 2 | Gemini 2.0 Flash (Preview) | 68.6 |
| 3 | GPT-4o | 65 |
| 4 | InternVL3-78B | 64.3 |
| 5 | InternVL3-38B | 62.9 |
| 6 | InternVL3-14B | 60.7 |
| 7 | Qwen 2 VL 72B | 59.3 |
| 8 | Claude 3.5 Sonnet (20240620) | 57.1 |
| 9 | Claude 3.5 Sonnet (20241022) | 52.1 |
| 10 | Ovis2-8B | 48.6 |
| 11 | Llama 4 Scout Base | 37.1 |
| 12 | Gemma 3 12B (IT) | 35.7 |
| 13 | Qwen 2 VL 7B | 35 |
| 14 | InternVL3-8B | 34.3 |
| 15 | Gemma 3 27B (IT) | 33.6 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-multilingual-ruozhiba-explanation-japanese · How the rankings work · Data refreshed daily, snapshot 2026-07-22.