List your company

Mathematical Reasoning on AIME24, AIME25, HMMT25

78.4AIME24 ScorePBSD
25.755239.422653.0966.7574May 6, 2026May 13, 2026May 21, 2026May 29, 2026Jun 6, 2026Jun 14, 2026Jun 22, 2026

Evaluation results

MethodLinks
PBSDModel=Qwen3-8B2026.0578.47146.165.2—
OPSDModel=Qwen3-8B2026.0577.370.145.164.2—
PBSDModel=Qwen3-4B2026.0577.36945.664—
SRPOModel=Qwen3-8B2026.0577.169.643.963.5—
OPSDModel=Qwen3-4B2026.0576.368.14663.5—
SRPOModel=Qwen3-4B2026.0576.367.845.563.2—
GRPOModel=Qwen3-8B2026.0576.16946.363.8—
DAPOModel=Qwen3-4B2026.0576.167.744.362.7—
DAPOModel=Qwen3-8B2026.057668.74663.6—
SDFTModel=Qwen3-8B2026.0575.969.144.263.1—
GRPOModel=Qwen3-4B2026.0575.767.844.462.6—
Base (Instruct)Model=Qwen3-8B2026.0575.665.343.661.5—
SDFTModel=Qwen3-4B2026.0575.667.144.762.5—
Base (Instruct)Model=Qwen3-4B2026.0574.566.14260.9—
SFTModel=Qwen3-8B2026.0571.963.942.459.4—
SFTModel=Qwen3-4B2026.0571.364.343.459.7—
PBSDModel=Qwen3-1.7B2026.0558.544.43044.3—
SRPOModel=Qwen3-1.7B2026.0558.144.129.443.9—
OPSDModel=Qwen3-1.7B2026.0557.143.329.343.2—
SDFTModel=Qwen3-1.7B2026.05574328.742.9—
Qwen3-1.7B w/ OPD+ReNIOTeacher Model=Qwen3-8B2026.0654.7243.0628.3342.04—
Qwen3-1.7B w/ OPDTeacher Model=Qwen3-8B2026.0654.4439.1627.540.37—
DAPOModel=Qwen3-1.7B2026.0551.538.623.537.9—
Base (Instruct)Model=Qwen3-1.7B2026.0551.437.223.137.3—
Qwen3-1.7BTeacher Model=Qwen3-8B2026.0651.333623.3336.89—
GRPOModel=Qwen3-1.7B2026.0550.638.423.537.5—
SFTModel=Qwen3-1.7B2026.0549.936.622.836.4—
DS-R1-Qwen-1.5B w/ OPD+ReNIOTeacher Model=DS-R1-Qwen-7B2026.0629.7223.8916.6723.43—
DS-R1-Qwen-1.5B w/ OPDTeacher Model=DS-R1-Qwen-7B2026.0628.8923.0614.4422.13—
DS-R1-Qwen-1.5BTeacher Model=DS-R1-Qwen-7B2026.0627.7822.7814.4421.67—
BaseModel=Qwen3-1.7B2026.05————37.4
BaseModel=Qwen3-4B2026.05————62.3
BaseModel=Qwen3-8B2026.05————62.1
Base+SkillModel=Qwen3-1.7B2026.05————35.9
Base+SkillModel=Qwen3-4B2026.05————59.9
Base+SkillModel=Qwen3-8B2026.05————62.4
GRPOModel=Qwen3-1.7B2026.05————37.5
GRPOModel=Qwen3-4B2026.05————62.7
GRPOModel=Qwen3-8B2026.05————65
GRPO+SkillModel=Qwen3-1.7B2026.05————39
GRPO+SkillModel=Qwen3-4B2026.05————61.3
GRPO+SkillModel=Qwen3-8B2026.05————63.4
OPSDModel=Qwen3-1.7B2026.05————42
OPSDModel=Qwen3-4B2026.05————63.5
OPSDModel=Qwen3-8B2026.05————66.8
OPSD+SkillModel=Qwen3-1.7B2026.05————40.6
OPSD+SkillModel=Qwen3-4B2026.05————62.2
OPSD+SkillModel=Qwen3-8B2026.05————62.3
SGSDModel=Qwen3-1.7B2026.05————43.7
SGSDModel=Qwen3-4B2026.05————64.3
SGSDModel=Qwen3-8B2026.05————65.5