List your company

Policy Evaluation on PolicyBench Level 3 CN

80.34AccuracyQwQ 32B
68.06871.25474.4477.626Apr 14, 2026

Evaluation results

MethodLinks
QwQ 32B2026.0480.34
o4-mini2026.0479.49
Gemini 2.5Model Variant=Gemini-2.5-Flash2026.0476.18
Claude 3.7Model Variant=Claude-3.7-Sonnet2026.0473.82
Gemini 2.0Model Variant=Gemini-2.0-Flash2026.0473.8
Deepseek R12026.0473.78
Claude 3.5Model Variant=Claude-3.5-Sonnet2026.0472.83
Deepseek V32026.0472.33
Gemma 3-27B2026.0471.51
GPT-4o2026.0470.24
LLaMA 42026.0468.54