Event ·Multi-source

GPT-6 Astra评测：规划出色但执行能力差

GPT-6 Astra被评价为目前最佳规划模型，能处理最复杂的技术任务，也是最好的代码审计工具。该模型在创意和写作方面有所提升，但执行能力仍然很差，无法构建简单的操作序列，代码错误率高。与Fable 5.1和Claude模型相比，GPT-6 Astra在执行质量上仍有明显差距。

Current read

Gary Marcus评测了GPT-6 Astra，发现它规划能力顶尖但执行糟糕，代码错误率高，远不如Fable和Claude平衡。

11 sources58° AI heatLast updated 2026/9/7 04:41:09

Evidence

11 sources

Primary sourceGary Marcus

2026/9/7 04:41:09View source

Related source 1lmarena.ai

2026/9/5 17:40:41View source

Related source 2shao__meng

2026/9/7 06:38:22View source

Related source 3Notion

2026/9/5 19:02:08View source

Related source 4宝玉

2026/9/5 23:13:44View source

Related source 5IT之家

2026/9/6 06:46:16View source

Related source 6AI Will

2026/9/8 00:28:36View source

Related source 7Blognone

2026/9/5 05:10:26View source

Related source 8向阳乔木

2026/9/5 06:19:44View source

Related source 9Decoder

2026/9/5 07:41:03View source

Related source 10Simon Willison

2026/9/5 15:05:32View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story