List your company

Mathematical Reasoning on MATH 500 (Accuracy @4, Average)

95.7Accuracy (avg@4)ASPO-Math-30B
58.98868.51978.0587.581Oct 7, 2025Nov 1, 2025Nov 27, 2025Dec 23, 2025Jan 17, 2026Feb 12, 2026Mar 10, 2026

Evaluation results

MethodLinks
ASPO-Math-30BBackbone=Qwen3-30B-A3B2025.1095.7—98.4
DS-R1-Distill-Qwen-7B + IC-DAPOBackbone=Qwen-7B, Zero-shot mode=true, Time/Step (s)=315.62026.0395.667.8—
DS-R1-Distill-Qwen-7B + CE-GPPOBackbone=Qwen-7B, Zero-shot mode=true, Time/Step (s)=292.52026.0395.367.3—
GRPOBackbone=Qwen3-30B-A3B2025.1095.3—97.8
DS-R1-Distill-Qwen-7B + DAPOBackbone=Qwen-7B, Zero-shot mode=true, Time/Step (s)=303.12026.0394.165.3—
DS-R1-Distill-Qwen-7B + GRPOBackbone=Qwen-7B, Zero-shot mode=true, Time/Step (s)=305.62026.0393.761.4—
DS-R1-Distill-Qwen-7BBackbone=Qwen-7B, Zero-shot mode=true2026.0393.661.8—
ASPO-Math-8BBackbone=Qwen3-8B2025.1093.6—96.4
ASPO-Math-4BBackbone=Qwen3-4B2025.1093.4—97
GRPOBackbone=Qwen3-8B2025.1092.7—96
GRPOBackbone=Qwen3-4B2025.1091.7—95.8
DS-R1-Distill-Qwen-1.5B + CE-GPPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=464.02026.039155.7—
DS-R1-Distill-Qwen-1.5B + IC-DAPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=477.22026.0390.656.4—
Nemotron-1.5BBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.1090.6—93.6
ASPO-Math-1.5BBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.1090.5—94.4
DS-R1-Distill-Qwen-1.5B + GSPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=437.32026.0390.355.7—
DS-R1-Distill-Qwen-1.5B + DAPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=459.62026.039053.9—
Qwen3-30B-A3BBackbone=Qwen3-30B-A3B2025.1088.4—96
DS-R1-Distill-Qwen-1.5B + GRPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=457.42026.0388.350.3—
DeepScaleR-1.5BBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.1087.7—93.6
GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.1087.6—94.6
DS-R1-Distill-Qwen-1.5BBackbone=Qwen-1.5B, Zero-shot mode=true2026.038646.3—
DS-R1-Distill-Qwen-1.5B + CISPOBackbone=Qwen-1.5B, Zero-shot mode=true, Time/Step (s)=466.32026.0385.848.8—
Qwen3-4BBackbone=Qwen3-4B2025.1084.5—92.4
DeepSeek-R1-1.5BBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.1083.6—92.4
Qwen3-8BBackbone=Qwen3-8B2025.1083.6—92.4
MCTS with Const-o-TBackbone=GPT-4.12025.1083.1——
MCTSBackbone=GPT-4.12025.1082.2——
CoVBackbone=GPT-4.12025.1081.7——
ToTBackbone=GPT-4.12025.1080.8——
MCTS with CoTBackbone=GPT-4.12025.1080.2——
CoTBackbone=GPT-4.12025.1079.2——
Const-o-TBackbone=GPT-4.12025.1079.2——
CoT + RSBackbone=GPT-4.12025.1078.4——
CTPBackbone=GPT-4.12025.1078.3——
MCTS with Const-o-TBackbone=LLaMA-32025.1070——
MCTS with CoTBackbone=LLaMA-32025.1068.8——
CoVBackbone=LLaMA-32025.1068.2——
CoT + RSBackbone=LLaMA-32025.1066.6——
MCTSBackbone=LLaMA-32025.1066.2——
CTPBackbone=LLaMA-32025.1065——
CoTBackbone=LLaMA-32025.1064.2——
Const-o-TBackbone=LLaMA-32025.1064.2——
ToTBackbone=LLaMA-32025.1060.4——