Cognition(@cognition_labs)
Try out today in Devin CLI and Devin Desktop: https://t.co/y3humFLPXs
6.0内容质量
TL;DR · AI 摘要
Grok 4.7在Devin中发布，但性能略逊于Grok 4.6，适用于复杂后端工程任务。
核心要点
- Grok 4.7在FrontierCode 1.1基准测试中得分为59.4%
- Devin CLI和Devin Desktop现已开放试用
- Grok 4.7在复杂任务中表现优异但存在过度扩展问题
结构提纲
按章节快速跳转。
宣布Grok 4.7在Devin平台上线，提供试用链接。
- ·性能数据
在FrontierCode 1.1基准测试中得分为59.4%。
- ›版本对比
Grok 4.7在复杂任务中表现优于Grok 4.6，但整体得分略低。
- ·产品试用
提供Devin CLI和Devin Desktop的下载链接。
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发布
- 性能数据
- FrontierCode 1.1得分59.4%
- 产品发布
- Devin CLI试用
- Devin Desktop试用
- 版本对比
- 优于Grok 4.6在部分任务
- 整体得分略低
金句 / Highlights
值得收藏与分享的关键句。
Grok 4.7在Extended测试中得分59.4%，表现优于Grok 4.6在部分硬核后端任务。
Grok 4.7在FrontierCode 1.1基准测试中得分略低于Grok 4.6，因过度扩展问题。
Devin CLI和Devin Desktop现已开放试用，提供产品下载链接。
#Devin#Grok#AI模型#工程工具
打开原文Cognition on X: "Try out today in Devin CLI and Devin Desktop: https://t.co/y3humFLPXs" / X
[](https://x.com/)
[](https://x.com/)
Cognition on X: "Try out today in Devin CLI and Devin Desktop: https://t.co/y3humFLPXs"
-  Cognition @cognition 10h Grok 4.7 is now available in Devin. On FrontierCode 1.1, Grok 4.7 scores 59.4% on Extended. In our evaluation, we see the model performs very well on hard backend engineering tasks.  [](https://x.com/cognition/status/2102147831519666639/photo/1) 36 30 447 73K
-  Cognition @cognition 10h 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: From devin.ai 4 4 33 8.1K
-  Cognition @cognition Try out today in Devin CLI and Devin Desktop: From devin.ai 9:27 PM · Sep 21, 2026·5,261 Views 1 1 9 1
-  369@X_com_Japan 9h free promotion？？ [](https://x.com/X_com_Japan/status/2102164028772806802) 37
See all the replies
Log in or sign up for X
See what’s happening and join the conversation
Continue with phoneContinue with Apple Continue with Google
or