MEGA-Bench Task - YouTube Video Info Parsing: leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | InternVL3-78B | 89.3 |
| 2 | Qwen 2 VL 72B | 83.3 |
| 3 | InternVL3-8B | 82.1 |
| 4 | GPT-4o | 81 |
| 5 | InternVL2.5-78B | 81 |
| 6 | Qwen 2 VL 7B | 76.2 |
| 7 | Gemini 2.5 Pro (Preview 03-25) | 76.2 |
| 8 | GPT-4o Mini | 75 |
| 9 | InternVL3-38B | 73.8 |
| 10 | Claude 3.5 Sonnet (20241022) | 72.6 |
| 11 | Llama 4 Scout Base | 72.6 |
| 12 | Claude 3.5 Sonnet (20240620) | 69 |
| 13 | Qwen 2 VL 2B | 59.5 |
| 14 | Gemini 1.5 Pro (002) | 52.4 |
| 15 | Gemma 3 27B (IT) | 50 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-youtube-video-info-parsing · How It Works · Data refreshed daily, snapshot 2026-09-05.