Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Evaluation on HVD-Implicit
Loading...
1.77
HCRate
GPT-4o
-9.9116
68.9392
147.79
226.6408
Jun 4, 2026
HCRate
NRRate
Updated 1mo ago
Evaluation Results
Method
Method
Links
HCRate
NRRate
GPT-4o
Attack=None
2026.06
1.77
74.49
Claude-3.7-Sonnet
Attack=None
2026.06
4.42
51.17
DeepSeek-V3
Attack=None
2026.06
20.35
87.11
GHOSTWRITER
Target Model=GPT-4o, A...
2026.06
53.1
83.68
GHOSTWRITER
Target Model=Claude-3....
2026.06
97.35
92.87
GHOSTWRITER
Target Model=DeepSeek-...
2026.06
147.79
92.73
GHOSTWRITER
Target Model=GPT-4o, A...
2026.06
175.22
84.38
GHOSTWRITER
Target Model=Claude-3....
2026.06
209.73
94.11
GHOSTWRITER
Target Model=DeepSeek-...
2026.06
293.81
92.33
Feedback
Search any
task
Search any
task