MEGA-Bench Task - Unusual Images — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 90.7 |
| 2 | Gemini 2.5 Pro | 89.3 |
| 3 | Claude 3.5 Sonnet (20240620) | 89.3 |
| 4 | Gemini 2.0 Flash (Preview) | 89 |
| 5 | Gemini 1.5 Pro (002) | 87.6 |
| 6 | GPT-4o Mini | 87.2 |
| 7 | Gemini 1.5 Flash (002) | 86.9 |
| 8 | Claude 3.5 Sonnet (20241022) | 83.4 |
| 9 | Gemma 3 27B (IT) | 80.7 |
| 10 | Gemma 3 12B (IT) | 80.7 |
| 11 | InternVL3-38B | 78.6 |
| 12 | InternVL3-78B | 77.9 |
| 13 | InternVL2.5-78B | 76.9 |
| 14 | Llama 4 Scout Base | 75.2 |
| 15 | Gemma 3 4B (IT) | 71 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-unusual-images · How the rankings work · Data refreshed daily, snapshot 2026-07-22.