List your company

Agent Safety Judgment on ROME CA unsafe subset

97.35F1 ScoreClaude Sonnet 4.6
89.591691.605893.6295.6342May 5, 2026

Evaluation results

MethodLinks
Claude Sonnet 4.6Model=Claude Sonnet 4.62026.0597.3594.8597
GPT-5.2Model=GPT-5.22026.0594.0588.7898
GPT-5Model=GPT-52026.0591.8985100
Claude Opus 4.6Model=Claude Opus 4.62026.0589.8981.6398