List your company

Mathematical Reasoning on GSM-Symbolic (Avg@4 accuracy)

88.12Avg@4 AccuracyReQueR
37.534450.667263.876.9328Apr 28, 2026

Evaluation results

MethodLinks
ReQueRModel=Qwen2.5-72B-Instruct, Temperature=0.62026.0488.12
ReQueRModel=Llama-3.1-70B-Instruct, Temperature=0.62026.0487.98
ReQueRModel=Qwen3-8B, Temperature=0.62026.0487.08
CoTModel=Qwen3-8B, Temperature=0.62026.0486.96
CoTModel=Llama-3.1-70B-Instruct, Temperature=0.62026.0485.76
CoTModel=Qwen2.5-72B-Instruct, Temperature=0.62026.0482.26
ReQueRModel=Qwen3-1.7B, Temperature=0.62026.0478.64
ReQueRModel=Llama-3.1-8B-Instruct, Temperature=0.62026.0475.82
CoTModel=Qwen3-1.7B, Temperature=0.62026.0474.62
CoTModel=Llama-3.1-8B-Instruct, Temperature=0.62026.0474.44
ReQueRModel=Llama-3.2-3B-Instruct, Temperature=0.62026.0469.38
ReQueRModel=Qwen3-0.6B, Temperature=0.62026.0466.14
CoTModel=Llama-3.2-3B-Instruct, Temperature=0.62026.0460.64
ReQueRModel=Mixtral-8x7B-Instruct, Temperature=0.62026.0459.66
CoTModel=Mixtral-8x7B-Instruct, Temperature=0.62026.0452.12
CoTModel=Qwen3-0.6B, Temperature=0.62026.0450.8
ReQueRModel=DS-MoE-16B-Chat, Temperature=0.62026.0448.34
CoTModel=DS-MoE-16B-Chat, Temperature=0.62026.0439.48