MATH (Mathematics Aptitude Test of Heuristics)

The original 12,500-problem competition-mathematics benchmark from Hendrycks et al. 2021; its 5,000-problem test split is the parent of the smaller MATH-500 subset most current model cards actually report.

Also known as: Hendrycks MATH, MATH dataset, hendrycks_math, minerva_math

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorymath
Subcategorycompetition mathematics
Page statusactive
Metricaccuracy (pass@1, boxed-answer match)
Directionhigher_is_better
Unit%
Dataset size5000
Dataset licenceMIT
PublisherUC Berkeley

What it measures

MATH gives a model a written competition mathematics problem -- spanning algebra, geometry, number theory, counting and probability, precalculus, intermediate algebra and prealgebra, each labelled with a difficulty level from 1 to 5 -- and asks for a full worked solution ending in a final answer. There are no answer choices. It tests the same skill as MATH-500 (this repository's `math_500` page, a fixed 500-problem subset of MATH's test split): multi-step symbolic and numeric reasoning, not speed, tool use, or any language beyond English. What differs between MATH and MATH-500 is size and provenance, not task design -- they test the identical kind of problem at 5,000 versus 500 items.

Task format

Free-response: read a competition mathematics problem, produce a worked solution and a final answer, conventionally inside \boxed{}.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub