DashBench - Q4 Event Detection: leaderboard
DashBench evaluates video understanding on 1,709 driving clips. Five explanation prompts and binary event detection are scored separately; mean agreement excludes the binary task. The 823-clip corner-case-window subset and training-free decoder variants retain their own conditions. Q5 asks for definition-guided corner-case identification and description. This board reports binary event-detection accuracy (%). Higher is better.
Metric: Binary event-detection accuracy (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | GPT-5 | 97.83 | #167 | |
| 2 | Gemini 3 Flash | 94.85 | #163 | |
| 3 | MiniCPM-V-4.5 | 87.53 | #1390 | |
| 4 | InternVL3.5-8B | Evaluation run: DashBench SECOND | 81.33 | #1558 |
| 5 | InternVL3.5-38B | 77.65 | #986 | |
| 6 | InternVL3.5-241B | 76.36 | #625 | |
| 7 | Qwen 3 VL 8B | 74.96 | #899 | |
| 8 | Qwen 3 VL 235B A22B | 72.03 | #484 | |
| 9 | Qwen 3 VL 32B | 71.04 | #609 | |
| 10 | Ovis2.5-9B | 64.47 | #1236 | |
| 11 | VITA-1.5 | 61.32 | #3196 | |
| 12 | InternLM-XComposer2.5 | 54.44 | #3981 | |
| 13 | LLaVA-OV-1.5-8B | 54.13 | #4771 | |
| 14 | InternVideo2.5-Chat-8B | 54 | #2017 | |
| 15 | GLM-4.1V-9B (Thinking) | 35.34 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=dashbench-q4-event-detection · How It Works · Data refreshed daily, snapshot 2026-11-16.