SaLAD - Unsafe Queries: leaderboard
Metric: Accuracy rate (%; GPT-4o judge; 1,070 hazardous daily-life queries whose answer must name and explain the specific risk; a bare refusal is wrong). Source: arxiv.org. Saturation forecast: Around December 2026. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.7 Sonnet | 57.2 |
| 2 | Gemini 2.5 Flash | 55.05 |
| 3 | GPT-4o | 53.83 |
| 4 | Llama 3.2 11B Vision | 33.46 |
| 5 | Qwen 2.5 VL 7B | 31.59 |
| 6 | MiniCPM-V-2.6 | 23.93 |
Interactive version: theaggregate.ai/benchmark?slug=salad-unsafe-queries · How It Works · Data refreshed daily, snapshot 2026-09-26.