List your company

Safety Evaluation on Sorry-Bench base

92.73Safety ScoreBase
21.105239.700158.29576.8899Jan 27, 2026

Evaluation results

MethodLinks
BaseLLM Model=Llama-3.3-70B2026.0192.73
AOALLM Model=Gemma-2-27B2026.0187.73
BaseLLM Model=Gemma-2-27B2026.0186.82
SkeletonLLM Model=Gemma-2-27B2026.0186.36
FormalLLM Model=Gemma-2-27B2026.0184.09
IDGAFLLM Model=Gemma-2-27B2026.0182.95
GCGLLM Model=Gemma-2-27B2026.0182.95
IDGAFLLM Model=Qwen2.5-32B2026.0177.27
AOALLM Model=Qwen2.5-32B2026.0174.77
AOALLM Model=Llama-3.3-70B2026.0167.73
GCGLLM Model=Qwen2.5-32B2026.0164.24
BaseLLM Model=Qwen2.5-32B2026.0164.09
FormalLLM Model=Qwen2.5-32B2026.0163.18
FormalLLM Model=Llama-3.3-70B2026.0162.73
SkeletonLLM Model=Qwen2.5-32B2026.0160.91
GCGLLM Model=Llama-3.3-70B2026.0156.82
Refusal SuppressionLLM Model=Qwen2.5-32B2026.0151.14
SkeletonLLM Model=Llama-3.3-70B2026.0143.41
Refusal SuppressionLLM Model=Gemma-2-27B2026.0142.27
FTLLM Model=Gemma-2-27B2026.0140.68
IDGAFLLM Model=Llama-3.3-70B2026.0140
FTLLM Model=Qwen2.5-32B2026.0138.18
FTLLM Model=Llama-3.3-70B2026.0137.95
Refusal SuppressionLLM Model=Llama-3.3-70B2026.0134.32
RULLM Model=Qwen2.5-32B2026.0128.64
RULLM Model=Gemma-2-27B2026.0124.32
RULLM Model=Llama-3.3-70B2026.0123.86