Skip to main content

leaderboard.cn

Explore

Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio

Library

Models
Pricing
Organizations
Benchmarks
Data sources

Loading page content

© 2026 Leaderboard.cn. Traceable AI model evaluations and rankings.

About Ranking methodology Changelog

v0.1.26.a85cb1563d03浙ICP备09076226号-7

Benchmark details loading

Back to benchmarks

MATHReasoning

Results for 30 models collected from 2 public sources.

#ModelScore

1

MiMo-V2.5-Pro BaseXiaomi

86.2%

2

Qwen3-235B-A22B-BaseAlibaba

71.8%

3

Kimi K2 BaseMoonshot AI

70.2%

4

MiMo-V2.5 BaseXiaomi

67.7%

5

DeepSeek-V4-Pro BaseDeepSeek

64.5%

6

Llama 4 Maverick BaseMeta

63.3%

7

Qwen2.5-Plus-BaseAlibaba

62.8%

8

DeepSeek-V3 BaseDeepSeek

62.6%

9

Qwen2.5-72BAlibaba

62.1%

10

Qwen3-14B-BaseAlibaba

62.0%

30 models total