MEGA-Bench Task - Video Eval Factual Pref — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 85.7 |
| 2 | Claude 3.5 Sonnet (20240620) | 78.6 |
| 3 | Gemini 2.0 Flash (Preview) | 78.6 |
| 4 | GPT-4o | 71.4 |
| 5 | Gemma 3 27B (IT) | 71.4 |
| 6 | Gemma 3 12B (IT) | 71.4 |
| 7 | InternVL3-8B | 71.4 |
| 8 | InternVL3-38B | 71.4 |
| 9 | MiniCPM-V-2.6 | 71.4 |
| 10 | InternVL2.5-78B | 71.4 |
| 11 | InternVL3-78B | 64.3 |
| 12 | Gemini 1.5 Pro (002) | 64.3 |
| 13 | Gemini 1.5 Flash (002) | 64.3 |
| 14 | Gemma 3 4B (IT) | 57.1 |
| 15 | InternVL2-8B | 57.1 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-video-eval-factual-pref · How the rankings work · Data refreshed daily, snapshot 2026-07-22.