List your company

Mathematical Reasoning on Track II Competition-Level Reasoning (test)

78.47Math ScoreGMPO
75.225276.067676.9177.7524Mar 10, 2026

Evaluation results

MethodLinks
GMPO2026.0378.4764.2737.2548.4219.589.586025.8642.93
GMPO+CLIPOReward Head=CLIPO2026.0377.8767.1438.8347.6719.3811.6761.2826.2443.76
GSPO2026.0377.5864.5638.3650.0816.2510.4260.1725.5842.87
GRPO+CLIPOReward Head=CLIPO2026.0377.4967.2138.9749.1718.1310.6361.2225.9743.6
DAPO+CLIPOReward Head=CLIPO2026.0377.4366.5539.8248.4218.7513.3361.2726.8344.05
GSPO+CLIPOReward Head=CLIPO2026.0377.1666.6937.6650.5320.639.3860.526.8443.67
GRPO2026.0376.4664.8337.8646.8417.929.5859.7224.7842.25
DAPO2026.0376.0366.0839.9447.3619.797.9260.6825.0242.85
Base Model2026.0375.3559.2435.3144.1310.636.2556.6320.3338.48