List your company

Mathematical Reasoning on (MATH500, AMC12, AIME24, AIME25, HMMT25) (test)

86MATH500 ScoreFailure-Prefix
83.484.07584.7585.425Jan 28, 2026

Evaluation results

MethodLinks
Failure-PrefixIteration=Iteration 2, Maximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.018658.334.725.715.644
Failure-PrefixCheckpoint=400, Maximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.0185.756.33325.916.243.4
Failure-PrefixIteration=Extended Iteration 1, Maximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.0185.156.333.425.516.543.4
MediumMaximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.018557.83226.21543.2
SaturateMaximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.0184.251.7302413.540.7
BaseMaximum token length=32000, Temperature=0.6, Top-p=1, Number of rollouts per question=322026.0183.553.128.42414.240.6