List your company

Attack Success Rate Evaluation on Hot100

26ASRSkillAttack
06.7513.520.25Apr 5, 2026

Evaluation results

MethodLinks
SkillAttackModel=glm-52026.0426
SkillAttackModel=gpt-5.42026.0423
SkillAttackModel=hunyuan-2.0-instruct-202511112026.0423
SkillAttackModel=doubao-seed-1-8-2512282026.0419
SkillAttackModel=gemini-3.0-pro-preview2026.0416
SkillAttackModel=kimi-k2.52026.0415
SkillAttackModel=gpt-5.4-nano2026.0412
SkillAttackModel=Qwen3.5-Plus2026.0411
SkillAttackModel=claude-sonnet-4-5-202509292026.0410
SkillAttackModel=MiniMax-M2.52026.049
Direct AttackModel=hunyuan-2.0-instruct-202511112026.044
Direct AttackModel=glm-52026.043
Direct AttackModel=gpt-5.42026.042
Direct AttackModel=kimi-k2.52026.042
Direct AttackModel=Qwen3.5-Plus2026.042
Direct AttackModel=MiniMax-M2.52026.042
Direct AttackModel=doubao-seed-1-8-2512282026.042
Direct AttackModel=gpt-5.4-nano2026.041
Direct AttackModel=claude-sonnet-4-5-202509292026.041
Direct AttackModel=gemini-3.0-pro-preview2026.041