TAR-Bench - Binary QA: leaderboard
Metric: Accuracy (%; yes/no questions about the anomaly, one positive and one negative per clip; zero-shot in non-reasoning (direct-answer) mode through VLMEvalKit, mean of 3 runs, on TAR-Bench: 960 human-curated annotations for 80 held-out traffic-anomaly clips from 17 public YouTube videos). Source: arxiv.org. Saturation forecast: Around 2029. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 27B (Non-reasoning) | 72.5 |
| 2 | Qwen 3.5 9B (Non-reasoning) | 67.3 |
| 3 | Qwen 3 VL 32B Instruct | 65 |
| 4 | Gemini 3.1 Pro (Preview) | 64.4 |
| 5 | Qwen 3 VL 8B Instruct | 59.6 |
| 6 | Gemma 4 31B (IT) | 56.5 |
Interactive version: theaggregate.ai/benchmark?slug=tar-bench-binary-qa · How It Works · Data refreshed daily, snapshot 2026-09-29.