Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Jailbreak Detection on ALL-4

92.6Accuracy

gpt-5-mini

52.903263.209173.51583.8209Feb 14, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
92.6
2026.02
88.89
2026.02
87.36
2026.02
85.89
2026.02
85.82
85
2026.02
83.95
2026.02
82.65
2026.02
81.05
72.9
2026.02
69.8
2026.02
54.64
2026.02
54.43