SafeSceneReason - Evidence Synthesis: leaderboard
Accuracy on the evidence-synthesis questions of SafeSceneReason (2026), an industrial-safety multiple-choice benchmark. Percent, higher is better.
Metric: Accuracy (%; the 101 evidence synthesis questions; single-answer multiple choice on the 1,000-question test set balanced across answer positions, built from workplace scene graphs and NIOSH FACE and CSB accident-report figures; temperature 0, unparseable answers counted wrong). Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 93.07 | #48 |
| 2 | GPT-5.6 Sol | 91.09 | #29 |
| 3 | Gemini 3.1 Pro (Preview) | 80.2 | #91 |
| 4 | Gemini 3.5 Flash | 76.24 | #98 |
| 5 | Kimi K2.6 | 73.27 | #173 |
| 6 | Claude Fable 5 | 73.27 | #26 |
| 7 | Qwen 3.6 27B | 42.57 | #339 |
| 8 | Qwen 3.5 9B | 36.63 | #803 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=safescenereason-evidence-synthesis · How It Works · Data refreshed daily, snapshot 2026-11-12.