Claude Sonnet 5 · FrontierCode · v1 · 成绩与配置

Claude Sonnet 5 的 FrontierCode · v1 记录。本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录),附 2 条来源定位。

成绩与评测条件

Claude Sonnet 5 · FrontierCode 成绩与配置
指标与成绩配置与评测条件出处
Score18.1%原文 18.1% · 单位 %
公布 未确认
Low effort (figure label)
设置
The figure labels this run as low effort; it does not provide a complete generation configuration.
未知项
Sampling parameters, token limits, and tool setup are not fully stated for the cited chart point.
工具
工具条件未记录
推理预算
low
框架
未记录
协议
FrontierCode v1; Figure 8.4.A plots score by effort setting and average cost per task.
条件
Low effort as labeled. The figure caption says costs were recomputed from Cognition's published per-task token splits at API list rates; other sampling and context settings are not stated.
在排名中查看此成绩
Score26.6%原文 26.6% · 单位 %
公布 未确认
Medium effort (figure label)
设置
The figure labels this run as med effort; it does not provide a complete generation configuration.
未知项
Sampling parameters, token limits, and tool setup are not fully stated for the cited chart point.
工具
工具条件未记录
推理预算
med
框架
未记录
协议
FrontierCode v1; Figure 8.4.A plots score by effort setting and average cost per task.
条件
Med effort as labeled. The figure caption says costs were recomputed from Cognition's published per-task token splits at API list rates; other sampling and context settings are not stated.
在排名中查看此成绩
Score28.9%原文 28.9% · 单位 %
公布 未确认
High effort (figure label)
设置
The figure labels this run as high effort; it does not provide a complete generation configuration.
未知项
Sampling parameters, token limits, and tool setup are not fully stated for the cited chart point.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
FrontierCode v1; Figure 8.4.A plots score by effort setting and average cost per task.
条件
High effort as labeled. The figure caption says costs were recomputed from Cognition's published per-task token splits at API list rates; other sampling and context settings are not stated.
在排名中查看此成绩
Score34%原文 34.0% · 单位 %
公布 未确认
Xhigh effort (figure label)
设置
The figure labels this run as xhigh effort; it does not provide a complete generation configuration.
未知项
Sampling parameters, token limits, and tool setup are not fully stated for the cited chart point.
工具
工具条件未记录
推理预算
xhigh
框架
未记录
协议
FrontierCode v1; Figure 8.4.A plots score by effort setting and average cost per task.
条件
Xhigh effort as labeled. The figure caption says costs were recomputed from Cognition's published per-task token splits at API list rates; other sampling and context settings are not stated.
在排名中查看此成绩
Score38.8%原文 38.8% · 单位 %
公布 未确认
Max effort (figure label)
设置
The figure labels this run as max effort; it does not provide a complete generation configuration.
未知项
Sampling parameters, token limits, and tool setup are not fully stated for the cited chart point.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
FrontierCode v1; Figure 8.4.A plots score by effort setting and average cost per task.
条件
Max effort as labeled. The figure caption says costs were recomputed from Cognition's published per-task token splits at API list rates; other sampling and context settings are not stated.
在排名中查看此成绩

出处与披露

测试版本与指标

别名 FrontierCode v1

  • Score · % · 越高越好