List your company

Policy Question Answering on PolicyBench Chinese

65.33AccuracyQwQ-32B
55.907658.353860.863.2462Apr 14, 2026

Evaluation results

MethodLinks
QwQ-32B2026.0465.33
Gemini-2.5-Flash2026.0463.6
Claude-3.7-sonnet2026.0463.19
Deepseek-R12026.0462.24
Claude-3.5-Sonnet2026.0462.11
o4-mini2026.0460.41
Gemini-2.0-Flash2026.0459.35
Deepseek-V32026.0458.82
LLaMA-42026.0458.3
GPT-4o2026.0457.53
Gemma 3-27B2026.0456.27