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.

语言EN
满分1
参评模型25
论文 ↗

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 ↗
5 Kimi K2 Instruct Moonshot AI 97.4 来源 ↗
6 Llama 3.1 Nemotron Ultra 253B v1 NVIDIA 97.0 来源 ↗
7 Llama-3.3 Nemotron Super 49B v1 NVIDIA 96.6 来源 ↗
8 Kimi-k1.5 Moonshot AI 96.2 来源 ↗
9 Claude 3.7 Sonnet Anthropic 96.2 来源 ↗
10 DeepSeek R1 Zero DeepSeek 95.9 来源 ↗
11 Llama 3.1 Nemotron Nano 8B V1 NVIDIA 95.4 来源 ↗
12 Phi 4 Mini Reasoning Microsoft 94.6 来源 ↗
13 DeepSeek R1 Distill Llama 70B DeepSeek 94.5 来源 ↗
14 DeepSeek R1 Distill Qwen 32B DeepSeek 94.3 来源 ↗
15 DeepSeek-V3 0324 DeepSeek 94.0 来源 ↗
16 DeepSeek R1 Distill Qwen 14B DeepSeek 93.9 来源 ↗
17 DeepSeek R1 Distill Qwen 7B DeepSeek 92.8 来源 ↗
18 QwQ-32B-Preview Alibaba Cloud / Qwen Team 90.6 来源 ↗
19 QwQ-32B Alibaba Cloud / Qwen Team 90.6 来源 ↗
20 DeepSeek-V3 DeepSeek 90.2 来源 ↗
21 o1-mini OpenAI 90.0 来源 ↗
22 DeepSeek R1 Distill Llama 8B DeepSeek 89.1 来源 ↗
23 DeepSeek R1 Distill Qwen 1.5B DeepSeek 83.9 来源 ↗
24 Granite 3.3 8B Base IBM 69.0 来源 ↗
25 Granite 3.3 8B Instruct IBM 69.0 来源 ↗