Início › Benchmarks › FrontierMath
FrontierMath
Benchmark
Capacidade
Research-grade mathematics — built unsolvable, already yielding.
Está a ver um registo de benchmark: um teste que a indústria usa para avaliar chips ou modelos — e o que ele realmente indica.
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.
Fontes
Registro revisado em 2026-01-15 — ver Metodologia.