Beta version: *Information might not be fully accurate. Please report any discrepancies.

MATH (CoT)

mathMathHigher is better

Challenging competition mathematics problems (AIME/IMO level).

Loading chart…

Loading leaderboard…

Benchmark Info

Current SOTA

Best Open Source

Median Score87.1%

DistributionBalanced

Score Distribution

P10

P50

P90

0100%

GSM8K

gsm8k

AIME 2024/25

aime

AIME 2026

aime-2026

AA Math Index

aa-math-index

HMMT Feb 2025

hmmt-feb-2025

MATH-500

math-500