List your company

Mathematical Reasoning on AIME Think (test)

72.4AccuracyQwen3-8B
60.64863.69966.7569.801Jan 11, 2026

Evaluation results

MethodLinks
Qwen3-8BTraining=RL, Mode=Mid-Think2026.0172.415,31844,142
Qwen3-8BTraining=RL, Mode=No-Think2026.017218,98034,585
Qwen3-8BTraining=No, Mode=Untrained2026.0171.616,84740,342
Qwen3-8BTraining=RL, Mode=Think2026.0169.813,33034,701
Qwen3-4BTraining=RL, Mode=Mid-Think2026.0169.615,79940,216
Qwen3-4BTraining=RL, Mode=No-Think2026.0169.320,55161,341
Qwen3-4BTraining=No, Mode=Untrained2026.0168.716,69940,089
Qwen3-4BTraining=RL, Mode=Think2026.0161.113,34732,118