Microsoft Phi-4 Reasoning: Small Model, Big Math
Microsoft's Phi-4 reasoning family delivers near-70B-class math performance in a 14B open-weight package, but the overthinking problem is real and the use case is narrower than the benchmarks suggest.
Aime
Best AI Models for Math Reasoning - March 2026
GPT-5.2 and Claude Opus 4.6 both score 100% on AIME 2025, while Gemini 3.1 Pro leads GPQA Diamond at 94.3% for PhD-level scientific reasoning.
Math Olympiad AI Leaderboard - March 2026 Rankings
Rankings of AI models on competition mathematics benchmarks including AIME 2025, IMO, MathArena, and FrontierMath, measuring the cutting edge of mathematical reasoning.
Reasoning Benchmarks: GPQA, AIME, and Humanity's Last Exam
Rankings of AI models on the hardest reasoning benchmarks available: GPQA Diamond, AIME competition math, and the notoriously difficult Humanity's Last Exam.
