MEGA-Bench Task - IconQA Count And Reasoning — leaderboard
Metric: Task Score (%). Source: huggingface.co. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 63.2 |
| 2 | Gemini 2.0 Flash (Preview) | 63.2 |
| 3 | InternVL3-78B | 52.6 |
| 4 | InternVL3-14B | 52.6 |
| 5 | InternVL3-38B | 47.4 |
| 6 | InternVL3-8B | 42.1 |
| 7 | Llama 4 Scout Base | 42.1 |
| 8 | Ovis2-8B | 31.6 |
| 9 | Gemma 3 27B (IT) | 26.3 |
| 10 | Gemma 3 12B (IT) | 15.8 |
| 11 | Gemma 3 4B (IT) | 15.8 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-iconqa-count-and-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.