MEGA-Bench Task - Multilingual Ruozhiba Explanation Arabic — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 73.6 |
| 2 | GPT-4o | 70.7 |
| 3 | Gemini 2.0 Flash (Preview) | 70 |
| 4 | InternVL3-38B | 62.9 |
| 5 | Claude 3.5 Sonnet (20241022) | 60.7 |
| 6 | Claude 3.5 Sonnet (20240620) | 57.9 |
| 7 | InternVL3-14B | 57.1 |
| 8 | InternVL2.5-78B | 55 |
| 9 | InternVL3-78B | 50.7 |
| 10 | Qwen 2 VL 72B | 46.4 |
| 11 | Ovis2-8B | 45 |
| 12 | InternVL3-8B | 42.9 |
| 13 | Gemma 3 12B (IT) | 42.1 |
| 14 | Llama 4 Scout Base | 37.1 |
| 15 | Gemma 3 27B (IT) | 30 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-multilingual-ruozhiba-explanation-arabic · How the rankings work · Data refreshed daily, snapshot 2026-07-22.