List your company

Jailbreak Defense on JailbreakBench

70Rate of Response SafetyDefaultSystem
0.663218.664136.66554.6659Jun 18, 2025

Evaluation results

MethodLinks
DefaultSystemLLM=Llama-2-7b-chat2025.067010030
SystemEmbedderLLM=Llama-2-7b-chat2025.0656.6710043.33
DefaultSystemLLM=Llama-3.1-8b2025.063010070
SystemEmbedderLLM=Llama-3.1-8b2025.063010070
LoRA*LLM=Mistral-7B-v0.22025.0623.3310076.67
SysformerLLM=Phi-3.5-mini2025.06209070
SysformerLLM=zephyr-7b-beta2025.0616.6793.3376.66
LoRA*LLM=Phi-3.5-mini2025.0616.676043.33
DefaultSystemLLM=Mistral-7B-v0.22025.0613.3383.3370
SystemEmbedderLLM=Mistral-7B-v0.22025.0613.3386.6773.34
LoRA*LLM=Llama-2-7b-chat2025.061096.6786.67
LoRA*LLM=Llama-3.1-8b2025.061096.6786.67
SysformerLLM=Mistral-7B-v0.22025.061010090
DefaultSystemLLM=zephyr-7b-beta2025.066.6733.3326.66
SystemEmbedderLLM=zephyr-7b-beta2025.066.674033.33
SysformerLLM=Llama-2-7b-chat2025.066.679083.33
LoRA*LLM=zephyr-7b-beta2025.063.3386.6773.33
SysformerLLM=Llama-3.1-8b2025.063.3396.6793.34
DefaultSystemLLM=Phi-3.5-mini2025.063.33106.67
SystemEmbedderLLM=Phi-3.5-mini2025.063.3316.6713.34