List your company
ResearchFormerly WizWand

Jailbreak on WildJailBreak (WJB) (test)

2ASR@1Jailbreak-R1
-1.5422.35546.2570.145May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Jailbreak-R1Target Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.052
AutoDAN-TurboTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.052.5
PAIRTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.054
CrescendoTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.057.5
MUSE-ATarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.058.5
MUSE-ATarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0514.73
CrescendoTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0522.5
PAIRTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0524
PAIRTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0531.5
Jailbreak-R1Target Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0531.5
X-TeamingTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0532.5
X-TeamingTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0534
ActorAttackTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0535.5
ActorAttackTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0538
MUSE-ATarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0540
ActorAttackTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0541.5
X-TeamingTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0545
Jailbreak-R1Target Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0549
AutoDAN-TurboTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0550.5
CrescendoTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0555.5
SirenTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0557.5
TROJailTarget Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0563.83
TROJailTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0566.83
SirenTarget Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0568
AutoDAN-TurboTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0573.5
SirenTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0575
TROJailTarget Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0577.8
TRACE (mix)Target Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0583.17
TRACE (single)Target Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0584
TRACE (single)Target Model=gpt-oss-20b, Evaluation Judge=HarmBench Classifier, Turns=52026.0584.17
TRACE (mix)Target Model=Llama3.1-8B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0588.67
TRACE (single)Target Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0589.83
TRACE (mix)Target Model=Qwen2.5-7B-IT, Evaluation Judge=HarmBench Classifier, Turns=52026.0590.5