FrontierMath

FrontierMath is a benchmark of over 1,000 original, computational math problems created to resist memorization, introduced in late 2024. It evaluates AI models' advanced reasoning by requiring novel problem-solving rather than reliance on training da

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.