MEGA-Bench - Skill: Mathematical And Logical Reasoning — leaderboard
Metric: Average Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 59.2 |
| 2 | Claude 3.5 Sonnet (20241022) | 48.9 |
| 3 | Gemini 2.0 Flash (Preview) | 47.8 |
| 4 | Claude 3.5 Sonnet (20240620) | 47.7 |
| 5 | GPT-4o | 44.2 |
| 6 | InternVL3-78B | 43.7 |
| 7 | Gemini 1.5 Pro (002) | 42.6 |
| 8 | InternVL3-38B | 40.4 |
| 9 | InternVL2.5-78B | 39.4 |
| 10 | Gemma 3 27B (IT) | 39.2 |
| 11 | Qwen 2 VL 72B | 39.1 |
| 12 | InternVL3-14B | 38.7 |
| 13 | Gemini 1.5 Flash (002) | 38 |
| 14 | InternVL3-8B | 34.8 |
| 15 | GPT-4o Mini | 34.1 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-skill-mathematical-and-logical-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.