Benchmark

MATH-500

math reasoning text

MATH-500 is a subset of the MATH dataset containing 500 challenging competition mathematics problems from AMC 10, AMC 12, AIME, and other mathematics competitions. Each problem includes full step-by-step solutions and spans multiple difficulty levels across seven mathematical subjects including Prealgebra, Algebra, Number Theory, Counting and Probability, Geometry, Intermediate Algebra, and Precalculus.

榜单来源

Top 10

榜单概览

1GLM-4.598.22GLM-4.5-Air98.13Nemotron Nano 9B v297.84Kimi K2-Instruct-090597.4
数据说明

分数为 0-100 百分制综合成绩（Elo 榜为相对评分，仅供同榜内比较）。评测分数属事实数据，整理自公开评测体系与厂商公开资料，部分分数为厂商自报口径，不同榜单的评测方法与数据集可能存在差异，请结合多个榜单交叉参考。

Ranking

模型排名

名次 模型 机构 分数 来源
1 GLM-4.5 Zhipu AI 98.2 来源
2 GLM-4.5-Air Zhipu AI 98.1 来源
3 Nemotron Nano 9B v2 NVIDIA 97.8 来源
4 Kimi K2-Instruct-0905 Moonshot AI 97.4 来源