List your company

Jailbreak Defense on Jailbreak Attack Suite

100AIM Defense RateJailbreak Antidote
-4235077Apr 16, 2026

Evaluation results

MethodLinks
Jailbreak AntidoteBase Model=Llama-3.1-8B-Instruct2026.0410097010086936377
SurgicalBase Model=Llama-3.1-8B-Instruct2026.0410076619888906782.86
AlphaSteerBase Model=Llama-3.1-8B-Instruct2026.04100965997899810091.29
BiPOBase Model=Llama-3.1-8B-Instruct2026.04100100100100949910099
FineSteerBase Model=Llama-3.1-8B-Instruct2026.0410010093100951009998.14
AlphaSteerBase Model=Qwen2.5-7B-Instruct2026.04100100979795959897.43
BiPOBase Model=Qwen2.5-7B-Instruct2026.04100100100100999910099.71
FineSteerBase Model=Qwen2.5-7B-Instruct2026.04100100981009710010099.28
AlphaSteerBase Model=Gemma-2-9B-IT2026.04100999999959410098
BiPOBase Model=Gemma-2-9B-IT2026.041001009610093989998
TruthFlowBase Model=Gemma-2-9B-IT2026.04100100921009910010098.71
FineSteerBase Model=Gemma-2-9B-IT2026.04100100100100—9510098.85
TruthFlowBase Model=Qwen2.5-7B-Instruct2026.049910090100971009997.85
TruthFlowBase Model=Llama-3.1-8B-Instruct2026.049690479886917383
Llama-3.1-8B-InstructBase Model=Llama-3.1-8B-Instruct2026.04924805875452849.43
CASTBase Model=Llama-3.1-8B-Instruct2026.049251679981969683.14
Jailbreak AntidoteBase Model=Qwen2.5-7B-Instruct2026.0491426905417347.14
CASTBase Model=Gemma-2-9B-IT2026.049174808366378073
SurgicalBase Model=Qwen2.5-7B-Instruct2026.0477816710079887071.71
Qwen2.5-7B-InstructBase Model=Qwen2.5-7B-Instruct2026.042521227119420.57
CASTBase Model=Qwen2.5-7B-Instruct2026.0425273396919910067.29
Jailbreak AntidoteBase Model=Gemma-2-9B-IT2026.0431144168473529.86
SurgicalBase Model=Gemma-2-9B-IT2026.042158875333629.14
Gemma-2-9B-ITBase Model=Gemma-2-9B-IT2026.04050756817824.71