List your company

Mathematical Reasoning on SVAMP (full set)

71.1Exact Match AccuracyGCE
36.57245.53654.563.464Jun 26, 2026

Evaluation results

MethodLinks
GCEModel=Qwen2.5, Params=1.5B2026.0671.1
SMMDModel=Qwen2.5, Params=1.5B2026.0671.1
CEModel=Qwen2.5, Params=1.5B2026.0668.6
NTILModel=Qwen2.5, Params=1.5B2026.0667.1
SMMDModel=SmolLM3, Params=3B2026.0667.1
CEModel=SmolLM3, Params=3B2026.0665.8
NTILModel=SmolLM3, Params=3B2026.0665.8
SMMDModel=Llama3, Params=8B2026.0665.3
GCEModel=SmolLM3, Params=3B2026.0663.7
CEModel=Llama3, Params=8B2026.0662.2
NTLModel=SmolLM3, Params=3B2026.0660.5
NTLModel=Qwen2.5, Params=1.5B2026.0659.8
NTILModel=Llama3, Params=8B2026.0659.5
GCEModel=Llama3, Params=8B2026.0658.4
NTLModel=Llama3, Params=8B2026.0657.5
SMMDModel=Qwen2.5, Params=0.5B2026.0643.9
NTLModel=Qwen2.5, Params=0.5B2026.0643.2
NTILModel=Qwen2.5, Params=0.5B2026.0640.6
CEModel=Qwen2.5, Params=0.5B2026.0639.8
GCEModel=Qwen2.5, Params=0.5B2026.0637.9