CSSBench - Privacy Leakage: leaderboard
Metric: Attack success rate (%; share of obfuscated malicious privacy leakage queries answered unsafely, Qwen3Guard-Gen-8B judge; lower is better). Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 8B | 5.33 |
| 2 | Qwen 3 0.6B | 6.67 |
| 3 | Qwen 3 1.7B | 14 |
Interactive version: theaggregate.ai/benchmark?slug=cssbench-privacy-leakage · How It Works · Data refreshed daily, snapshot 2026-09-26.