MEGA-Bench Task - Multiple States Identify Asia — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 91.4 |
| 2 | Gemini 2.0 Flash (Preview) | 90 |
| 3 | Ovis2-8B | 71.4 |
| 4 | Gemini 1.5 Pro (002) | 62.9 |
| 5 | Claude 3.5 Sonnet (20240620) | 58.6 |
| 6 | GPT-4o | 57.1 |
| 7 | Claude 3.5 Sonnet (20241022) | 57.1 |
| 8 | Gemini 1.5 Flash (002) | 47.1 |
| 9 | GPT-4o Mini | 41.4 |
| 10 | Gemma 3 27B (IT) | 37.1 |
| 11 | Gemma 3 12B (IT) | 30 |
| 12 | InternVL3-78B | 28.6 |
| 13 | InternVL3-38B | 22.9 |
| 14 | InternVL3-14B | 21.4 |
| 15 | InternVL2.5-78B | 21.4 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-multiple-states-identify-asia · How the rankings work · Data refreshed daily, snapshot 2026-07-22.