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

Jailbreak Detection on Average of six attacks

0Avg Success Rate

GradSafe

-3.7621.624772.38Dec 3, 2025Dec 22, 2025Jan 11, 2026Jan 31, 2026Feb 20, 2026Mar 12, 2026Apr 1, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2025.12
0-
2025.12
0-
2025.12
0-
2025.12
2-
2025.12
5-
2025.12
5-
2025.12
5-
2025.12
6-
2025.12
6-
2025.12
7-
2026.04
7.4518.5
2026.04
8.9529.96
2025.12
12-
2026.04
13.6836.41
2025.12
19-
2025.12
22-
2026.04
23.5146.53
2025.12
27-
2025.12
28-
2025.12
30-
2025.12
31-
2026.04
35.7777.73
2026.04
36.0574.86
2026.04
40.6184
2025.12
50-
2025.12
52-
2025.12
54-
2026.04
54.94-
2025.12
59-
2025.12
64-
2025.12
73-
2025.12
74-
2025.12
74-
2025.12
75-
2025.12
81-
2025.12
85-
2025.12
93-
2025.12
94-