List your company

Mathematical Problem Solving on MATH (pass@50)

90pass@50GPT-4o-mini
71.07275.98680.985.814Oct 4, 2025

Evaluation results

MethodLinks
GPT-4o-miniSampling Strategy=GUIDEDSAMPLING2025.10905
GPT-4o-miniSampling Strategy=Repeated Sampling2025.1085.713.2
Phi-4-mini-instructSampling Strategy=GUIDEDSAMPLING2025.1080.83.4
Phi-4-mini-instructSampling Strategy=Repeated Sampling2025.1071.82.1