List your company

Mathematical Problem Solving on OlymMATH-EN

48.3AccuracyKACE
40.39642.44844.546.552May 30, 2026

Evaluation results

MethodLinks
KACESolver model=gpt-4.1-mini2026.0548.3
Tiered Best-of-NSolver model=gpt-4.1-mini2026.0544.4
Tiered + GEPASolver model=gpt-4.1-mini2026.0544.4
Best-of-5 self-consistencySolver model=gpt-4.1-mini2026.0540.7
Tiered + ACESolver model=gpt-4.1-mini2026.0540.7