A3Bench - Compound Questions: leaderboard
Zero-shot multiple-choice video QA accuracy (%) on multi-attribute compound role-action questions in A3Bench (arXiv:2611.24762).
Metric: Compound Questions Accuracy (%). Source: arxiv.org. Saturation forecast: Around 2029. 14 models tracked. Updated 20 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 46.65 | #1035 |
| 2 | InternVL2.5-8B | 45.01 | #2370 |
| 3 | Gemma 4 26B A4B (IT) | 39.65 | #561 |
| 4 | InternVideo2.5-8B | 39.34 | |
| 5 | GPT-5.1 | 39.17 | #242 |
| 6 | InternVL3-9B | 39.06 | #1731 |
| 7 | Ovis2.5-9B | 37.99 | #1189 |
| 8 | Qwen 2.5 VL 7B Instruct | 37.15 | #1904 |
| 9 | InternVL3.5-8B | 37.03 | #1599 |
| 10 | Qwen 3 VL 8B (Thinking) | 36.41 | #920 |
| 11 | VideoLLaMA3-7B | 30.86 | #2000 |
| 12 | ViCLIP | 21.61 | |
| 13 | LLaVA-Video-7B-Qwen2 | 19.34 | #1977 |
| 14 | X-CLIP | 15.75 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=a3bench-compound-questions · How It Works · Data refreshed daily, snapshot 2026-11-20.