MEGA-Bench Task - Trance Physics Reasoning Basic: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 03-25) | 100 |
| 2 | GPT-4o | 88.2 |
| 3 | InternVL3-38B | 88.2 |
| 4 | InternVL2.5-78B | 88.2 |
| 5 | Claude 3.5 Sonnet (20240620) | 88.2 |
| 6 | InternVL3-78B | 82.4 |
| 7 | InternVL3-8B | 82.4 |
| 8 | Gemini 1.5 Pro (002) | 76.5 |
| 9 | Qwen 2 VL 72B | 70.6 |
| 10 | Claude 3.5 Sonnet (20241022) | 70.6 |
| 11 | GPT-4o Mini | 70.6 |
| 12 | Gemma 3 27B (IT) | 64.7 |
| 13 | Gemma 3 12B (IT) | 47.1 |
| 14 | Pixtral-12B | 47.1 |
| 15 | Gemini 1.5 Flash (002) | 47.1 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-trance-physics-reasoning-basic · How It Works · Data refreshed daily, snapshot 2026-09-05.