Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agent Safety Judgment on ROME CA unsafe subset
Loading...
97.35
F1 Score
Claude Sonnet 4.6
89.5916
91.6058
93.62
95.6342
May 5, 2026
F1 Score
Recall
Validity Score
Updated 28d ago
Evaluation Results
Method
Method
Links
F1 Score
Recall
Validity Score
Claude Sonnet 4.6
Model=Claude Sonnet 4.6
2026.05
97.35
94.85
97
GPT-5.2
Model=GPT-5.2
2026.05
94.05
88.78
98
GPT-5
Model=GPT-5
2026.05
91.89
85
100
Claude Opus 4.6
Model=Claude Opus 4.6
2026.05
89.89
81.63
98
Feedback
Search any
task
Search any
task