Benchmark

PolyMATH

math reasoning spatial_reasoning multimodal vision multimodal

Polymath is a challenging multi-modal mathematical reasoning benchmark designed to evaluate the general cognitive reasoning abilities of Multi-modal Large Language Models (MLLMs). The benchmark comprises 5,000 manually collected high-quality images of cognitive textual and visual challenges across 10 distinct categories, including pattern recognition, spatial reasoning, and relative reasoning.

榜单来源

Top 10

榜单概览

1Qwen3-235B-A22B-Thinkin…60.12Qwen3-Next-80B-A3B-Thin…56.33Qwen3-Next-80B-A3B-Inst…45.9
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 Qwen3-235B-A22B-Thinking-2507 Alibaba Cloud / Qwen Team 60.1 来源
2 Qwen3-Next-80B-A3B-Thinking Alibaba Cloud / Qwen Team 56.3 来源
3 Qwen3-Next-80B-A3B-Instruct Alibaba Cloud / Qwen Team 45.9 来源