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

Safety Evaluation on StrongReject

0.64Attack Success Rate

DirectRefusal

-396.77442,285.77284,968.327,650.8672May 30, 2025Jul 21, 2025Sep 12, 2025Nov 3, 2025Dec 26, 2025Feb 16, 2026Apr 10, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
0.64--
2026.04
1.3--
2026.04
3.5--
2026.01
3.51--
2026.03
5.4--
2025.05
62-
2026.04
6--
2025.05
82.09-
2025.05
101.99-
2026.01
11.82--
2025.05
122.13-
2025.05
142.24-
2025.05
152.23-
2025.05
152.31-
2026.03
16.3--
2025.05
182.46-
2026.03
25.6--
2026.01
26.52--
2026.04
30.5-0.245
2026.01
32--
2026.01
33.55--
2026.01
33.87--
2026.04
39.6--
2026.01
41.85--
2026.04
44-0.38
2026.03
45.2--
2026.01
46.33--
2026.04
46.9-0.073
2026.01
47.92--
2026.04
49.8-0.102
2026.01
51.44--
2026.03
51.8--
2026.01
53.99--
2026.01
54.63--
2026.01
63.58--
2026.01
64--
2026.01
64.86--
2026.03
66--
2026.01
73.16--
2026.01
74.12--
2026.01
76.68--
2026.01
77--
2026.01
82.11--
2026.04
82.4--
2026.01
82.43--
2026.01
82.75--
2026.04
83.1--
2026.04
85--
2026.04
86.6--
2026.04
87.2--
2026.04
88.2--
2026.01
160--
2026.01
767--
2026.01
831--
2026.01
1,661--
2026.01
2,971--
2026.01
3,387--
2026.01
7,188--
2026.01
8,083--
2026.01
8,243--
2026.01
8,466--
2026.01
9,042--
2026.01
9,105--
2026.01
9,744--
2026.01
9,936--
2025.03
-0.4054-
2025.03
-0.379-
2025.03
-0.4698-
2025.03
-0.5054-
2025.03
-0.7264-
2025.03
-0.4633-
2025.03
-0.8798-
2025.03
-0.8105-
2025.03
-0.8917-
2025.03
-0.9361-
2025.03
-0.3808-
2025.03
-0.3792-
2025.03
-0.4952-
2025.03
-0.5026-
2025.03
-0.5577-
2025.03
-0.5062-
2025.03
-0.8486-
2025.03
-0.5155-
2025.03
-0.596-
2025.03
-0.8798-
2026.04
---2.6
2026.04
---8
2026.04
---87.2
2026.04
--31.9
2026.04
--35.4
2026.04
--56.3