MEGA-Bench Task - Generated Video Artifacts — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 49.4 |
| 2 | InternVL3-14B | 46.2 |
| 3 | Pixtral-12B | 43.8 |
| 4 | GPT-4o Mini | 43.7 |
| 5 | InternVL3-8B | 40.6 |
| 6 | Gemma 3 27B (IT) | 40 |
| 7 | Llama 4 Scout Base | 40 |
| 8 | Gemma 3 12B (IT) | 36.9 |
| 9 | Gemini 1.5 Pro (002) | 36.9 |
| 10 | Gemma 3 4B (IT) | 36.3 |
| 11 | Gemini 2.0 Flash (Preview) | 36.2 |
| 12 | Gemini 1.5 Flash (002) | 35.6 |
| 13 | GPT-4o | 35 |
| 14 | InternVL3-38B | 33.1 |
| 15 | Claude 3.5 Sonnet (20240620) | 32.5 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-generated-video-artifacts · How the rankings work · Data refreshed daily, snapshot 2026-07-22.