List your company

Mathematical Reasoning on AIME 23-26

719Pass@1DFT
-28.76165.37359.5553.63Jul 6, 2026

Evaluation results

MethodLinks
DFTModel=Qwen3-14B2026.0771925
LP-SFTModel=Qwen3-4B2026.0768235.83
ASFTModel=Qwen3-14B2026.0761725
EAFTModel=Qwen3-4B2026.0759428.33
DFTModel=Qwen3-4B2026.0758321.67
CEModel=Qwen3-4B2026.0757827.5
LP-SFTModel=Qwen3-14B2026.0756830.83
CEModel=Qwen3-14B2026.0753624.17
GEMModel=Qwen3-4B2026.0753429.17
GEMModel=Qwen3-14B2026.0747122.5
EAFTModel=Qwen3-14B2026.0745122.5
ASFTModel=Qwen3-4B2026.0728918.33
LP-SFTBackbone=Qwen3-14B2026.071037.5
GEMModel=Llama3.1-8B2026.07103.33
CEBackbone=Qwen3-14B2026.078.9833.33
ASFTBackbone=Qwen3-14B2026.078.8527.5
EAFTBackbone=Qwen3-14B2026.078.731.67
GEMBackbone=Qwen3-14B2026.078.1235
LP-SFTBackbone=Qwen3-4B2026.077.2929.17
ASFTBackbone=Qwen3-4B2026.076.9828.33
DFTBackbone=Qwen3-14B2026.076.2815.83
DFTBackbone=Qwen3-4B2026.075.2613.33
CEModel=Llama3.1-8B2026.0751.67
ASFTModel=Llama3.1-8B2026.0751.67
LP-SFTModel=Llama3.1-8B2026.0751.67
GEMBackbone=Qwen3-4B2026.074.9530
EAFTBackbone=Qwen3-4B2026.074.8729.17
CEBackbone=Qwen3-4B2026.074.6125
EAFTModel=Llama3.1-8B2026.0730.83
ASFTBackbone=Qwen3-14B2026.072.9221.67
LP-SFTBackbone=Qwen3-14B2026.072.7619.17
LP-SFTBackbone=Qwen3-4B2026.072.4720.83
CEBackbone=Qwen3-14B2026.072.1615.83
EAFTBackbone=Qwen3-14B2026.071.9817.5
GEMBackbone=Qwen3-14B2026.071.9515.83
EAFTBackbone=Qwen3-4B2026.071.8215
CEBackbone=Qwen3-4B2026.071.815
ASFTBackbone=Qwen3-4B2026.071.5413.33
DFTBackbone=Qwen3-14B2026.071.388.33
GEMBackbone=Qwen3-4B2026.071.3515
EAFTBackbone=Llama3.1-8B2026.070.7314.17
CEBackbone=Llama3.1-8B2026.070.710
GEMBackbone=Llama3.1-8B2026.070.6512.5
DFTBackbone=Qwen3-4B2026.070.624.17
ASFTBackbone=Llama3.1-8B2026.070.527.5
LP-SFTBackbone=Llama3.1-8B2026.070.426.67
CEBackbone=Llama3.1-8B2026.070.266.67
DFTBackbone=Llama3.1-8B2026.070.265.83
EAFTBackbone=Llama3.1-8B2026.070.185.83
GEMBackbone=Llama3.1-8B2026.070.185
DFTBackbone=Llama3.1-8B2026.070.163.33
LP-SFTBackbone=Llama3.1-8B2026.070.13.33
ASFTBackbone=Llama3.1-8B2026.070.051.67
DFTModel=Llama3.1-8B2026.0700