List your company

Mathematical Reasoning on AIME Out-of-Distribution

69.1Initial F1AutoPyVerifier
32.742.1551.661.05Apr 24, 2026

Evaluation results

MethodLinks
AutoPyVerifierVerifier Model=GPT-5.4 mini2026.0469.173.34.2
AutoPyVerifierVerifier Model=Gemini-3.1 Pro2026.0466.177.311.2
AutoPyVerifierVerifier Model=GPT-5.42026.0450.174.824.7
AutoPyVerifierVerifier Model=Gemini-3.1 Fl2026.0441.291.650.4
AutoPyVerifierVerifier Model=Claude Hai-4.52026.0434.153.319.2