List your company

Competition-style Mathematical Reasoning on AIME 25

95.7ScoreGemini-3-Pro
79.68483.8428892.158Feb 6, 2026

Evaluation results

MethodLinks
Gemini-3-ProEvaluation protocol=Avg@102026.0295.7
GPT-5Evaluation protocol=Avg@102026.0294.3
Claude-Opus-4.5Evaluation protocol=Avg@102026.0291.3
Claude-Opus-4.1Evaluation protocol=Avg@102026.0280.3