List your company

Math Reasoning on AMC23, AMC24, AIME24, AIME25, MATH500

93.1AMC23 Scoreµ-GRPO
19.05238.27657.576.724May 17, 2026

Evaluation results

MethodLinks
µ-GRPOBackbone Model=DEEPSEEK-7B, µ (Staleness level)=10242026.0593.168.950.237.992.868.621.451.3
GRPOBackbone Model=DEEPSEEK-7B, µ (Staleness level)=42026.0591.466.746.833.292.466.12661.9
M2POBackbone Model=DEEPSEEK-7B, µ (Staleness level)=10242026.0590.669.450.8359367.827.968.3
GRPOBackbone Model=DEEPSEEK-7B, µ (Staleness level)=10242026.0582.663.539.526.888.760.222.354.5
µ-GRPOBackbone Model=QWEN3-BASE-8B, µ (Staleness level)=10242026.0569.451.121.517.481.848.27.925.9
GRPOBackbone Model=QWEN3-BASE-8B, µ (Staleness level)=42026.0568.853.323.519.486.350.321.358.2
GRPOBackbone Model=QWEN2.5-MATH-7B, µ (Staleness level)=42026.0568.142.82915.88147.38.823.7
M2POBackbone Model=QWEN2.5-MATH-7B, µ (Staleness level)=10242026.0566.940.629.614.682.546.86.720.2
M2POBackbone Model=QWEN3-BASE-8B, µ (Staleness level)=10242026.0565.653.924.219.687.150.123.249.7
µ-GRPOBackbone Model=QWEN2.5-MATH-7B, µ (Staleness level)=10242026.0563.147.231.516.581.7484.718.1
GRPOBackbone Model=QWEN2.5-MATH-7B, µ (Staleness level)=10242026.0561.336.223.39.878.241.83.817.9
GRPOBackbone Model=QWEN3-BASE-8B, µ (Staleness level)=10242026.0559.643.217.611.274.641.27.423.6
GRPOBackbone Model=QWEN3-BASE-1.7B, µ (Staleness level)=42026.0542.527.27.53.866.629.511.220.5
µ-GRPOBackbone Model=QWEN3-BASE-1.7B, µ (Staleness level)=10242026.0540.824.97.56.568.729.75.212.4
M2POBackbone Model=QWEN3-BASE-1.7B, µ (Staleness level)=10242026.053524.4104.466.628.112.924.2
GRPOBackbone Model=QWEN3-BASE-1.7B, µ (Staleness level)=10242026.0532.519.37.74.563.525.56.513.3
M2POBackbone Model=LLAMA-3.2-3B-INSTRUCT, µ (Staleness level)=10242026.0530.61510.60.250.721.415.321.3
µ-GRPOBackbone Model=LLAMA-3.2-3B-INSTRUCT, µ (Staleness level)=10242026.0529.417.7140.552.822.912.718.6
GRPOBackbone Model=LLAMA-3.2-3B-INSTRUCT, µ (Staleness level)=42026.052520.612.30.255.122.615.122.7
GRPOBackbone Model=LLAMA-3.2-3B-INSTRUCT, µ (Staleness level)=10242026.0521.912.29.80.646.818.313.219.3