List your company

Mathematical Reasoning on OlympiadBench (Acc, GT-pass)

53.4AccuracyBASE
30.72836.61442.548.386Jun 14, 2026

Evaluation results

MethodLinks
BASESolver=Gemini 3 Pro, K=82026.0653.499
BASESolver=R1-Qwen3-8B, K=82026.0653.2100
Independent FormalizationSolver=R1-Qwen3-8B, K=82026.0642.879.4
Independent FormalizationSolver=Gemini 3 Pro, K=82026.0641.777.6
BASESolver=Qwen3-8B, K=82026.0639.798.1
Independent FormalizationSolver=Qwen3-8B, K=82026.0631.680.4