FrontierMath
Benchmark
Kemampuan
Research-grade mathematics — built unsolvable, already yielding.
Anda sedang melihat entri benchmark: sebuah uji yang digunakan industri untuk mengukur chip atau model — dan apa yang sebenarnya ditunjukkannya.
Epoch AI commissioned original problems from working mathematicians (Fields medalists among the reviewers) that take experts hours or days. Launch scores were ~2%; frontier systems passed 25–40% on the easier tiers within a year — while a disclosure that OpenAI funded the set sharpened norms about benchmark independence.
Sumber
Rekaman ditinjau per 2026-01-15 — lihat Metodologi.