List your company

Jailbreak Attack on AdvBench GPT-4 1.0 (test)

0.77ASRMAC
-2.87621.734546.34570.9555Dec 29, 2025

Evaluation results

MethodLinks
MACAttack Type=Automated, Evaluator=GPT-42025.120.77
COLD-AttackAttack Type=Automated, Evaluator=GPT-42025.120.77
Base64Attack Type=Automated, Evaluator=GPT-42025.120.77
PromptAttackAttack Type=Manual, Evaluator=GPT-42025.120.96
GCGAttack Type=Automated, Evaluator=GPT-42025.121.73
ArtPromoptAttack Type=Manual, Evaluator=GPT-42025.121.75
CodeChameleonAttack Type=Manual, Evaluator=GPT-42025.1222.27
AutoDANAttack Type=Automated, Evaluator=GPT-42025.1226.54
PAIRAttack Type=Automated, Evaluator=GPT-42025.1227.18
DeepInceptionAttack Type=Manual, Evaluator=GPT-42025.1227.27
TAPAttack Type=Automated, Evaluator=GPT-42025.1240.97
SelfCipherAttack Type=Manual, Evaluator=GPT-42025.1241.73
GPTFuzzerAttack Type=Automated, Evaluator=GPT-42025.1242.5
ReNeLLMAttack Type=Manual, Evaluator=GPT-42025.1268.08
DRAAttack Type=Manual, Evaluator=GPT-42025.1282.88
FlipattackAttack Type=Manual, Evaluator=GPT-42025.1286.73
EquaCodeEvaluator=GPT-42025.1291.92