Benchmark

SuperGPQA

reasoning general math legal healthcare finance chemistry economics physics text

SuperGPQA is a comprehensive benchmark that evaluates large language models across 285 graduate-level academic disciplines. The benchmark contains 25,957 questions covering 13 broad disciplinary areas including Engineering, Medicine, Science, and Law, with specialized fields in light industry, agriculture, and service-oriented domains. It employs a Human-LLM collaborative filtering mechanism with over 80 expert annotators to create challenging questions that assess graduate-level knowledge and reasoning capabilities.

Top 10

榜单概览

1Qwen3-235B-A22B-Thinkin…64.92Qwen3-Next-80B-A3B-Thin…60.83Qwen3-Next-80B-A3B-Inst…58.84Kimi K2-Instruct-090557.2
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 64.9 来源
2 Qwen3-Next-80B-A3B-Thinking Alibaba Cloud / Qwen Team 60.8 来源
3 Qwen3-Next-80B-A3B-Instruct Alibaba Cloud / Qwen Team 58.8 来源
4 Kimi K2-Instruct-0905 Moonshot AI 57.2 来源