Jailbreak Attack
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
96.3Attack Success Rate (ASR)
4
Apr 21, 2026
90.5Attack Success Rate
4
Apr 21, 2026
1Bypass Rate
4
Apr 14, 2026
48.5Attack Success Rate
4
Mar 17, 2026
70Attack Success Rate
4
Mar 17, 2026
15Attack Success Rate
4
Mar 17, 2026
100Attack Success Rate (ASR)
4
Mar 4, 2026
10.5ASR
4
Feb 26, 2026
37ASR
4
Feb 26, 2026
54ASR
4
Feb 26, 2026
2.4Average Latency (s)
4
Feb 26, 2026
83.73Illegal Activity Success Rate
4
Feb 26, 2026
0.702IA (Illegal Activity)
4
Feb 26, 2026
32Attack Success Rate (ASR)
3
Jul 3, 2026
43.2ASR
3
Jul 3, 2026
93ASR
3
Apr 14, 2026
57Attack Success Rate (ASR)
3
Apr 14, 2026
91Attack Success Rate (ASR)
3
Apr 14, 2026
2.53Average Toxicity
3
Apr 13, 2026
2.38Average Toxicity (AvgTox)
3
Apr 13, 2026
2.8Average Toxicity
3
Apr 13, 2026
98.33ASR
3
Feb 26, 2026
99.17ASR
3
Feb 26, 2026
98.33ASR
3
Feb 26, 2026
90ASR (Refusal Matching)
3
Feb 26, 2026