List your company

Mathematical Reasoning on MATH500 (Pass@1, #Tk)

62Pass@1 RateDRP
12.91225.65638.451.144Jan 26, 2026

Evaluation results

MethodLinks
DRPBackbone=Phi-4-Reasoning2026.01622,122
DRPBackbone=GPT-oss-20B2026.0161.464,137
FROSTBackbone=Phi-4-Reasoning2026.0159.8344.37
TALEBackbone=Phi-4-Reasoning2026.01581,874.43
FROSTBackbone=GPT-oss-20B2026.0158680.89
BaseBackbone=Phi-4-Reasoning2026.0154.81,721.95
TALEBackbone=GPT-oss-20B2026.0154.543,878.87
ThinkLessBackbone=Phi-4-Reasoning2026.0154.141,101.21
ThinkLessBackbone=GPT-oss-20B2026.0154.12,206
BaseBackbone=GPT-oss-20B2026.01541,575.36
SelfBudgeterBackbone=Phi-4-Reasoning2026.0153.471,195.18
SelfBudgeterBackbone=GPT-oss-20B2026.0153.42,285
TALEBackbone=Magistral-Small-1.12026.0130.4723.91
FROSTBackbone=Magistral-Small-1.12026.0130.498.2
ThinkLessBackbone=Magistral-Small-1.12026.01251,405
SelfBudgeterBackbone=Magistral-Small-1.12026.01231,520
DRPBackbone=Magistral-Small-1.12026.01211,680.33
BaseBackbone=Magistral-Small-1.12026.0114.81,389.89