List your company

Agent Safety Judgment on ROME IR unsafe

95.74F1 ScoreClaude Sonnet 4.6
82.53285.96189.3992.819May 5, 2026

Evaluation results

MethodLinks
Claude Sonnet 4.6Model=Claude Sonnet 4.62026.0595.7491.8498
GPT-5.2Model=GPT-5.22026.0591.384100
Claude Opus 4.6Model=Claude Opus 4.62026.0586.0675.5394
GPT-5Model=GPT-52026.0583.0471100