A3S-Bench - Agent-Specific Operators - General Security Score: leaderboard
Metric: General Security Score (1-5 judge rating of the full trace; 5 = safe). Source: arxiv.org. Saturation forecast: Around May 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.4 | 4.52 |
| 2 | GLM-5.1 | 4.49 |
| 3 | Claude Sonnet 4.5 | 4.35 |
| 4 | GPT-5.2 | 4.07 |
| 5 | Kimi K2.5 | 3.55 |
| 6 | Qwen 3.5 397B A17B | 3.51 |
| 7 | DeepSeek V4 Flash | 3.38 |
| 8 | MiniMax-M2.5 | 3.3 |
| 9 | DeepSeek V3.2 | 3.23 |
| 10 | Qwen 3.5 122B A10B | 3.08 |
| 11 | Qwen 3.5 35B A3B | 2.81 |
Interactive version: theaggregate.ai/benchmark?slug=a3s-bench-agent-specific-operators-general-security-score · How It Works · Data refreshed daily, snapshot 2026-09-25.