Cognition(@cognition_labs)

On FrontierCode 1.1, our benchmark for real-world engineering tasks, Grok 4.7 scores slightly below ...

6.0内容质量
On FrontierCode 1.1, our benchmark for real-world engineering tasks, Grok 4.7 scores slightly below ...

TL;DR · AI 摘要

Grok 4.7在工程任务基准测试中得分低于Grok 4.6,存在过度扩展问题。

核心要点

  • Grok 4.7在FrontierCode 1.1基准中得分为59.4%
  • 模型在复杂后端任务表现优异但存在过度扩展问题
  • Grok 4.7已集成到Devin平台

结构提纲

按章节快速跳转。

  1. Grok 4.7在FrontierCode 1.1基准中得分低于Grok 4.6。

  2. 模型在复杂任务表现优异但存在过度扩展问题。

  3. Grok 4.7已集成到Devin平台并提供下载。

思维导图

用一张图看清主题之间的关系。

查看大纲文本(无障碍 / 无 JS 友好)
  • Grok 4.7测试分析
    • 测试结果
      • FrontierCode 1.1得分59.4%
    • 性能特征
      • 复杂任务表现优异
      • 过度扩展问题
    • 应用信息
      • 集成到Devin平台

金句 / Highlights

值得收藏与分享的关键句。

#AI模型#基准测试#工程任务
打开原文

Cognition on X: "On FrontierCode 1.1, our benchmark for real-world engineering tasks, Grok 4.7 scores slightly below Grok 4.6. While strong on many hard tasks, it tends to over scope on others, leading it to trail behind Grok 4.6 in the aggregate. Read more: https://t.co/PgginMwMNp" / X

[](https://x.com/)

[](https://x.com/)

Cognition on X: "On FrontierCode 1.1, our benchmark for real-world engineering tasks, Grok 4.7 scores slightly below Grok 4.6. While strong on many hard tasks, it tends to over scope on others, leading it to trail behind Grok 4.6 in the aggregate. Read more: https://t.co/PgginMwMNp"

Log in or sign up for X

See what’s happening and join the conversation

Continue with phoneContinue with Apple Continue with Google

or

Log in with username or email