Find by task
Math & science
Competition math, proofs, expert-level questions. Below: our own tests first, then public leaderboards with their current leaders.
Public leaderboards
3 from labs and universities.
-
FrontierMath
Hundreds of brand-new, unpublished math problems written by professional mathematicians, ranging from hard competition level up to problems that take experts days. Because the problems are secret, it is a strong check of real mathematical ability rather than memorization.
Leader: Claude Fable 5.2 (max) 95.1
-
MathArena
Tests AI models on fresh, hard math problems taken from new competitions and research papers, so the answers cannot have been memorised.
Leader: Gemini 4 Carbon (high) 94.28% ± 3.64%
-
PutnamBench
Tests whether AI systems can write machine-checkable formal proofs for problems from the Putnam, a famously hard undergraduate math competition.
Leader: 1 672