List your company

Jailbreak Attack on AdvBench GPT-4o 1.0 (test)

92.67ASRCodeChameleon
-3.706821.314146.33571.3559Dec 29, 2025

Evaluation results

MethodLinks
CodeChameleonAttack Type=Manual, Evaluator=GPT-42025.1292.67
FlipattackAttack Type=Manual, Evaluator=GPT-42025.1290.77
DRAAttack Type=Manual, Evaluator=GPT-42025.1288.84
EquaCodeEvaluator=GPT-42025.1287.12
ReNeLLMAttack Type=Manual, Evaluator=GPT-42025.1285.38
GPTFuzzerAttack Type=Automated, Evaluator=GPT-42025.1266.73
TAPAttack Type=Automated, Evaluator=GPT-42025.1261.63
Base64Attack Type=Automated, Evaluator=GPT-42025.1257.88
PAIRAttack Type=Automated, Evaluator=GPT-42025.1247.83
AutoDANAttack Type=Automated, Evaluator=GPT-42025.1246.92
DeepInceptionAttack Type=Manual, Evaluator=GPT-42025.1240.04
ArtPromoptAttack Type=Manual, Evaluator=GPT-42025.124.42
PromptAttackAttack Type=Manual, Evaluator=GPT-42025.121.92
GCGAttack Type=Automated, Evaluator=GPT-42025.121.15
MACAttack Type=Automated, Evaluator=GPT-42025.120.58
COLD-AttackAttack Type=Automated, Evaluator=GPT-42025.120.19
SelfCipherAttack Type=Manual, Evaluator=GPT-42025.120