List your company

Mathematical Reasoning on AIME No-think (test)

62.9AccuracyQwen3-8B
15.78828.01940.2552.481Jan 11, 2026

Evaluation results

MethodLinks
Qwen3-8BTraining=RL, Mode=No-Think2026.0162.913,59111,936
Qwen3-4BTraining=RL, Mode=No-Think2026.0136.717,42120,818
Qwen3-8BTraining=RL, Mode=Mid-Think2026.0127.67,1146,055
Qwen3-4BTraining=RL, Mode=Mid-Think2026.0124.912,49211,271
Qwen3-8BTraining=No, Mode=Untrained2026.0121.34,520447
Qwen3-4BTraining=No, Mode=Untrained2026.01204,605542
Qwen3-8BTraining=RL, Mode=Think2026.0119.37,6249,423
Qwen3-4BTraining=RL, Mode=Think2026.0117.67,973917