MEGA-Bench - Application: Knowledge — leaderboard
Metric: Average Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 69.4 |
| 2 | Gemini 2.0 Flash (Preview) | 62.5 |
| 3 | GPT-4o | 61.4 |
| 4 | Gemini 1.5 Pro (002) | 57.2 |
| 5 | Claude 3.5 Sonnet (20241022) | 56.6 |
| 6 | Claude 3.5 Sonnet (20240620) | 55.1 |
| 7 | GPT-4o Mini | 54 |
| 8 | InternVL3-78B | 51.6 |
| 9 | Gemini 1.5 Flash (002) | 51.3 |
| 10 | InternVL3-38B | 50.8 |
| 11 | Qwen 2 VL 72B | 50.5 |
| 12 | InternVL2.5-78B | 50.3 |
| 13 | Gemma 3 27B (IT) | 49.4 |
| 14 | InternVL3-14B | 47.8 |
| 15 | Gemma 3 12B (IT) | 47.4 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-application-knowledge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.