MEGA-Bench Task - NExT-QA OE — leaderboard
Metric: Task Score (%). Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 47.4 |
| 2 | GPT-4o | 44.2 |
| 3 | Gemini 2.0 Flash (Preview) | 41.6 |
| 4 | InternVL2.5-78B | 40 |
| 5 | InternVL3-78B | 37.4 |
| 6 | InternVL3-14B | 36.8 |
| 7 | Claude 3.5 Sonnet (20240620) | 36.8 |
| 8 | Claude 3.5 Sonnet (20241022) | 36.3 |
| 9 | Aquila-VL-2B | 35.8 |
| 10 | Llama 4 Scout Base | 35.8 |
| 11 | Gemma 3 27B (IT) | 35.3 |
| 12 | Gemini 1.5 Pro (002) | 35.3 |
| 13 | Pixtral-12B | 35.3 |
| 14 | InternVL3-8B | 34.7 |
| 15 | Gemma 3 12B (IT) | 34.2 |
Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-next-qa-oe · How the rankings work · Data refreshed daily, snapshot 2026-07-22.