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

Jailbreak Defense on StrongReject

1.5Attack Success Rate

R2D

-1.14416.70334.5552.397Mar 18, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.03
1.5----
2026.03
2.9----
2026.03
4.5----
2026.03
4.8----
2026.03
5.4----
2026.03
5.8----
2026.03
7----
2026.03
7.1----
2026.03
8.1----
2026.03
8.4----
2026.03
9.4----
2026.03
16----
2026.03
16.3----
2026.03
18.8----
2026.03
19.9----
2026.03
20.4----
2026.03
21.1----
2026.03
21.2----
2026.03
22.1----
2026.03
22.4----
2026.03
24.4----
2026.03
24.4----
2026.03
26----
2026.03
26.4----
2026.03
26.9----
2026.03
31.4----
2026.03
37.5----
2026.03
38.8----
2026.03
39.1----
2026.03
40.7----
2026.03
41.3----
2026.03
43.6----
2026.03
43.9----
2026.03
44.9----
2026.03
45.2----
2026.03
46.6----
2026.03
46.9----
2026.03
47.1----
2026.03
48.7----
2026.03
48.9----
2026.03
50.6----
2026.03
51.3----
2026.03
51.6----
2026.03
52.2----
2026.03
52.9----
2026.03
52.9----
2026.03
55.8----
2026.03
59.9----
2026.03
62.2----
2026.03
63.2----
2026.03
64.7----
2026.03
66----
2026.03
67----
2026.03
67.6----
2025.06
-2092.5572.55-
2025.06
-6.6731.9125.24-
2025.06
-6.6734.0427.37-
2025.06
-13.3375.5362.2-
2025.06
-1010090-
2025.06
-66.6798.9432.27-
2025.06
-6.674033.33-
2025.06
-3.3380.8577.52-
2025.06
-0100100-
2025.06
-3010070-
2025.06
-3010070-
2025.06
-3.3310096.67-
2025.06
-6.6748.9442.27-
2025.06
-3.3321.2817.95-
2025.06
-6.6726.619.93-
2025.06
-6.6758.5151.84-
2025.06
-1010090-
2025.06
-13.3393.6280.29-
2025.06
-13.3393.6280.29-
2025.06
-1010090-
2026.04
----3.5
2026.04
-----40.1
2026.04
-----146.4
2026.04
----24.8
2026.04
----48
2026.04
----75