List your company

Mathematical Problem Solving on AIME 2026 (Score %)

100AIME 2026 ScoreSTAR-PólyaMath
86.16889.75993.3596.941May 19, 2026May 26, 2026Jun 2, 2026Jun 9, 2026Jun 16, 2026Jun 23, 2026Jun 30, 2026

Evaluation results

MethodLinks
STAR-PólyaMathBackbone=all GPT-5.5xh2026.05100
GPT-5.4Reasoning Effort=xhigh2026.0599.17
GPT-5.2Reasoning Effort=high2026.0598.33
Gemini 3.1 Pro2026.0598.33
GPT-5.5Reasoning Effort=xhigh2026.0597.5
Claude Opus 4.6Reasoning Effort=high2026.0596.67
Claude Opus 4.7Reasoning Effort=xhigh2026.0595.83
DeepSeek-v4-ProReasoning Effort=Max2026.0595.83
Kimi K2.6Reasoning Effort=Think2026.0595.83
Kimi K2.5Reasoning Effort=Think2026.0595.83
GLM 5.12026.0595.83
Gemini-3-FlashTier=High2026.0693.3
GPT-5-miniTier=High2026.0692.5
Gemini 3 Pro2026.0591.67
Seed2.0 Lite2026.0688.3
Seed2.0 Mini2026.0686.7