跳到正文
智迹闻AuraTracer

发布于

GLM-5.2

开始使用

API 接入

Zhipu AI API文档
目录部署标识:glm-5.2。具体区域、权限与价格以服务商为准。

公开成绩与计算依据

综合参考分根据已公开的评测资料整理。查看发布日期依据

综合参考分 13.0 · 7项基准

推理求解 16.7

GPQA Diamond · 91.2

huggingface.co · 官方发布比较 · Diamond

GLM-5.2 (max) · Official published comparison; GPQA Diamond; exact column GLM-5.2 (max). Footnotes and all competing runs retained in source.

工具与任务执行 18.3

APEX-Agents · 35.6

huggingface.co · 官方发布比较 · current published task suite

GLM-5.2 (max) · Official published comparison; APEX-Agents; exact column GLM-5.2 (max). Footnotes and all competing runs retained in source.

AutomationBench · 12.9

huggingface.co · 官方发布比较 · current published task suite

GLM-5.2 (max) · Published page kimi-k3; row AutomationBench; preserve page footnotes and task-specific settings

AA-Briefcase · 0.3606060606060606 比例

Artificial Analysis · 独立评测 · 1.1

GLM-5.2 (Max) · aa-briefcase-rubric-reviewed-method-20261029

MCP Atlas · 82.6

huggingface.co · 官方发布比较 · 1000 tasks

GLM-5.2 (max) · Official published comparison; MCP-Atlas; exact column GLM-5.2 (max). Footnotes and all competing runs retained in source.

编程 0.0

DeepSWE · 46.2

huggingface.co · 官方发布比较 · 1.1

GLM-5.2 (max) · Published page kimi-k3; row DeepSWE; preserve page footnotes and task-specific settings

语言与信息处理 14.7

SimpleQA Verified · 0.342 比例

Epoch AI · 独立评测 · run 1.2.0

glm-5.2_max · epoch-simpleqa-reviewed-method-20261029-task-version-1.2.0 · 2026-08-27公开

查看原始成绩与计算数据

资料来源

模型资料来源 · 目录许可