List your company
ResearchFormerly WizWand

Jailbreak Attack on AdvBench GPT-3.5-turbo 1.0 (test)

97.12Attack Success RateEquaCode
-3.884822.337648.5674.7824Dec 20, 2025Jan 4, 2026Jan 20, 2026Feb 5, 2026Feb 20, 2026Mar 8, 2026Mar 24, 2026
Updated 7mo ago

Evaluation results

MethodLinks
EquaCodeEvaluator=GPT-42025.1297.12
TriageFuzzQuery Budget=252026.0394
DRAAttack Type=Manual, Evaluator=GPT-42025.1293.65
ReNeLLMAttack Type=Manual, Evaluator=GPT-42025.1291.35
HPM2025.1290.8
FlipattackAttack Type=Manual, Evaluator=GPT-42025.1288.65
TriageFuzzQuery Budget=202026.0388.5
PAP2025.1287.8
CodeChameleonAttack Type=Manual, Evaluator=GPT-42025.1284.62
AutoDANAttack Type=Automated, Evaluator=GPT-42025.1281.73
TriageFuzzQuery Budget=152026.0381
PAIRQuery Budget=252026.0374
CoA2025.1270.9
PAIRQuery Budget=202026.0368
TriageFuzzQuery Budget=102026.0368
TAPAttack Type=Automated, Evaluator=GPT-42025.1260.54
PAIRQuery Budget=152026.0360.5
PAIRAttack Type=Automated, Evaluator=GPT-42025.1259.68
TAPQuery Budget=252026.0358
PAIR2025.1255.8
TAPQuery Budget=202026.0354.5
GPTFuzzQuery Budget=252026.0352
PAIRQuery Budget=102026.0349.5
TAPQuery Budget=152026.0348.5
GPTFuzzQuery Budget=202026.0345.5
Base64Attack Type=Automated, Evaluator=GPT-42025.1245
GCGAttack Type=Automated, Evaluator=GPT-42025.1242.88
GPTFuzzQuery Budget=152026.0342
DeepInceptionAttack Type=Manual, Evaluator=GPT-42025.1241.13
TAPQuery Budget=102026.0340.5
AutoDAN2025.1240.2
GPTFuzzerAttack Type=Automated, Evaluator=GPT-42025.1237.79
MACAttack Type=Automated, Evaluator=GPT-42025.1236.15
COLD-AttackAttack Type=Automated, Evaluator=GPT-42025.1234.23
GPTFuzzQuery Budget=102026.0332.5
ArtPromoptAttack Type=Manual, Evaluator=GPT-42025.1214.06
PromptAttackAttack Type=Manual, Evaluator=GPT-42025.1213.46
SelfCipherAttack Type=Manual, Evaluator=GPT-42025.120