List your company

Scientific Reasoning on GPQA Diamond (Avg@4)

52.02Avg@4 AccuracyReQueR
23.659231.022138.38545.7479Apr 28, 2026

Evaluation results

MethodLinks
ReQueRModel=Qwen2.5-72B-Instruct, Temperature=0.62026.0452.02
CoTModel=Qwen2.5-72B-Instruct, Temperature=0.62026.0450.88
ReQueRModel=Llama-3.1-70B-Instruct, Temperature=0.62026.0448.36
CoTModel=Llama-3.1-70B-Instruct, Temperature=0.62026.0447.73
ReQueRModel=Qwen3-8B, Temperature=0.62026.0445.45
CoTModel=Qwen3-8B, Temperature=0.62026.0441.92
ReQueRModel=Qwen3-1.7B, Temperature=0.62026.0432.83
ReQueRModel=Llama-3.1-8B-Instruct, Temperature=0.62026.0431.31
CoTModel=Qwen3-1.7B, Temperature=0.62026.0431.31
ReQueRModel=Mixtral-8x7B-Instruct, Temperature=0.62026.0431.31
ReQueRModel=Llama-3.2-3B-Instruct, Temperature=0.62026.0430.3
ReQueRModel=Qwen3-0.6B, Temperature=0.62026.0430.3
CoTModel=Llama-3.1-8B-Instruct, Temperature=0.62026.0427.78
CoTModel=Qwen3-0.6B, Temperature=0.62026.0427.53
CoTModel=Llama-3.2-3B-Instruct, Temperature=0.62026.0426.77
ReQueRModel=DS-MoE-16B-Chat, Temperature=0.62026.0426.77
CoTModel=Mixtral-8x7B-Instruct, Temperature=0.62026.0425.76
CoTModel=DS-MoE-16B-Chat, Temperature=0.62026.0424.75