MEGA-Bench Task - Paper Review Writing — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 70.7 |
| 2 | Gemini 1.5 Flash (002) | 62 |
| 3 | InternVL3-8B | 61.3 |
| 4 | Claude 3.5 Sonnet (20241022) | 61.3 |
| 5 | Gemini 1.5 Pro (002) | 61.3 |
| 6 | InternVL3-78B | 60.7 |
| 7 | Gemma 3 27B (IT) | 60 |
| 8 | Gemma 3 12B (IT) | 59.3 |
| 9 | InternVL3-14B | 59.3 |
| 10 | Gemini 2.0 Flash (Preview) | 58.7 |
| 11 | Claude 3.5 Sonnet (20240620) | 58 |
| 12 | InternVL3-38B | 57.3 |
| 13 | GPT-4o | 55.3 |
| 14 | Qwen 2 VL 72B | 55.3 |
| 15 | Pixtral-12B | 55.3 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-paper-review-writing · How the rankings work · Data refreshed daily, snapshot 2026-07-22.