List your company

Jailbreak Attack on RMCBench (ASR, MR)

85.53ASRCodeSpear
5.148426.016746.88567.7533Jun 10, 2026

Evaluation results

MethodLinks
CodeSpearTarget Model=MiniMax-M2.72026.0685.5364.29
CodeSpearTarget Model=MiniMax-M2.72026.0685.5364.29
CodeSpearTarget Model=MiniMax-M2.52026.0684.6256.23
CodeSpearTarget Model=MiniMax-M2.52026.0684.6256.23
CodeSpear-TTarget Model=GPT-52026.0670.353.65
CodeJailbreakerTarget Model=GPT-OSS-120B2026.0668.6837.55
CodeJailbreakerTarget Model=GPT-OSS-120B2026.0668.6837.55
CodeSpearTarget Model=GPT-OSS-120B2026.0666.331.87
CodeSpearTarget Model=GPT-OSS-120B2026.0666.331.87
CodeSpear-TTarget Model=GPT-5-mini2026.0663.7544.58
CodeSpearTarget Model=GPT-52026.0655.4943.96
CodeSpearTarget Model=GPT-52026.0655.4943.96
CodeSpearTarget Model=GPT-5-mini2026.0653.4839.56
CodeSpearTarget Model=GPT-5-mini2026.0653.4839.56
CodeJailbreakerTarget Model=GPT-52026.0653.333.33
CodeJailbreakerTarget Model=GPT-52026.0653.333.33
APTTarget Model=GPT-52026.0640.1133.7
CodeJailbreakerTarget Model=MiniMax-M2.72026.0640.1125.64
APTTarget Model=GPT-52026.0640.1133.7
CodeJailbreakerTarget Model=MiniMax-M2.72026.0640.1125.64
APTTarget Model=GPT-5-mini2026.063731.14
APTTarget Model=GPT-5-mini2026.063731.14
PAIRTarget Model=GPT-5-mini2026.0636.4530.95
PAIRTarget Model=GPT-5-mini2026.0636.4530.95
CodeJailbreakerTarget Model=MiniMax-M2.52026.0636.2628.57
CodeJailbreakerTarget Model=MiniMax-M2.52026.0636.2628.57
APTTarget Model=GPT-OSS-120B2026.0635.921.98
APTTarget Model=GPT-OSS-120B2026.0635.921.98
Vanilla-TTarget Model=GPT-5-mini2026.0632.3330
Vanilla-TTarget Model=GPT-5-mini2026.0632.3330
VanillaTarget Model=GPT-52026.0632.0526.01
VanillaTarget Model=GPT-52026.0632.0526.01
LRLTarget Model=MiniMax-M2.52026.0631.1418.68
LRLTarget Model=MiniMax-M2.52026.0631.1418.68
LRLTarget Model=GPT-52026.0630.7727.11
LRLTarget Model=GPT-52026.0630.7727.11
LRLTarget Model=GPT-5-mini2026.0629.320.51
LRLTarget Model=GPT-5-mini2026.0629.320.51
PAIRTarget Model=GPT-52026.0629.1226.37
PAIRTarget Model=GPT-52026.0629.1226.37
CodeJailbreakerTarget Model=GPT-5-mini2026.0628.7525.27
CodeJailbreakerTarget Model=GPT-5-mini2026.0628.7525.27
VanillaTarget Model=GPT-5-mini2026.0627.8426.92
VanillaTarget Model=GPT-5-mini2026.0627.8426.92
Vanilla-TTarget Model=GPT-52026.0627.6727
Vanilla-TTarget Model=GPT-52026.0627.6727
DANTarget Model=GPT-52026.0624.9122.53
DANTarget Model=GPT-52026.0624.9122.53
PAIRTarget Model=GPT-OSS-120B2026.0623.6317.03
APTTarget Model=MiniMax-M2.72026.0623.6317.58
PAIRTarget Model=GPT-OSS-120B2026.0623.6317.03
APTTarget Model=MiniMax-M2.72026.0623.6317.58
VanillaTarget Model=MiniMax-M2.72026.0620.3317.4
VanillaTarget Model=MiniMax-M2.72026.0620.3317.4
LRLTarget Model=GPT-OSS-120B2026.0618.6813.19
LRLTarget Model=GPT-OSS-120B2026.0618.6813.19
LRLTarget Model=MiniMax-M2.72026.0617.9512.82
LRLTarget Model=MiniMax-M2.72026.0617.9512.82
PAIRTarget Model=MiniMax-M2.72026.0617.2212.09
PAIRTarget Model=MiniMax-M2.72026.0617.2212.09
APTTarget Model=MiniMax-M2.52026.0616.1212.82
APTTarget Model=MiniMax-M2.52026.0616.1212.82
PAIRTarget Model=MiniMax-M2.52026.0615.7511.9
PAIRTarget Model=MiniMax-M2.52026.0615.7511.9
VanillaTarget Model=MiniMax-M2.52026.0615.3813.37
VanillaTarget Model=MiniMax-M2.52026.0615.3813.37
VanillaTarget Model=GPT-OSS-120B2026.0614.8413.74
VanillaTarget Model=GPT-OSS-120B2026.0614.8413.74
Vanilla-TTarget Model=GPT-OSS-120B2026.0613.6713
Vanilla-TTarget Model=GPT-OSS-120B2026.0613.6713
DANTarget Model=MiniMax-M2.52026.0612.458.24
DANTarget Model=MiniMax-M2.52026.0612.458.24
Vanilla-TTarget Model=MiniMax-M2.72026.0611.338.67
Vanilla-TTarget Model=MiniMax-M2.72026.0611.338.67
DANTarget Model=GPT-5-mini2026.0610.999.52
DANTarget Model=MiniMax-M2.72026.0610.999.89
DANTarget Model=GPT-5-mini2026.0610.999.52
DANTarget Model=MiniMax-M2.72026.0610.999.89
Vanilla-TTarget Model=MiniMax-M2.52026.0610.339
Vanilla-TTarget Model=MiniMax-M2.52026.0610.339
DANTarget Model=GPT-OSS-120B2026.068.247.33
DANTarget Model=GPT-OSS-120B2026.068.247.33