MEGA-Bench Task - Rebus — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 69.6 |
| 2 | Gemini 2.5 Pro | 60.9 |
| 3 | Claude 3.5 Sonnet (20241022) | 52.2 |
| 4 | Claude 3.5 Sonnet (20240620) | 52.2 |
| 5 | Gemini 2.0 Flash (Preview) | 43.5 |
| 6 | Gemma 3 27B (IT) | 39.1 |
| 7 | Gemini 1.5 Pro (002) | 39.1 |
| 8 | Gemma 3 12B (IT) | 34.8 |
| 9 | GPT-4o Mini | 30.4 |
| 10 | Gemini 1.5 Flash (002) | 30.4 |
| 11 | InternVL3-78B | 26.1 |
| 12 | InternVL2.5-78B | 26.1 |
| 13 | Llama 4 Scout Base | 26.1 |
| 14 | Gemma 3 4B (IT) | 21.7 |
| 15 | InternVL3-8B | 17.4 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-rebus · How the rankings work · Data refreshed daily, snapshot 2026-07-22.