Event ·Multi-source

GPT-6 Astra基准测试结果不一

OpenAI的GPT-6 Astra在基准测试中表现不一。Epoch AI将其评为169分，排名第一；而Artificial Analysis则认为其表现与前任相当，落后于Claude Fable 5.1。最引人注目的是在ARC-AGI-3测试中，Astra首次以超越普通人类的效率完成测试。ARC Prize负责人François Chollet虽不认为这已证明AGI实现，但他认为进步速度比预期快一倍，并提前了AGI预测时间。

Current read

GPT-6 Astra在ARC-AGI-3测试中首次超越人类效率，让AI专家Chollet提前了AGI预测时间。

11 sources73° AI heatLast updated 2026/9/4 11:07:36

Evidence

11 sources

Primary sourceDecoder

2026/9/4 11:07:36View source

Related source 1orange.ai

2026/9/3 23:36:56View source

Related source 2向阳乔木

2026/9/4 08:57:07View source

Related source 3lmarena.ai

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

Related source 4The Rundown AI

2026/9/3 18:07:36View source

Related source 5Simon Willison’s Weblog

2026/9/3 20:18:41View source

Related source 6歸藏(guizang.ai)

2026/9/4 00:59:38View source

Related source 7Gary Marcus

2026/9/4 01:34:52View source

Related source 8IT之家

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

Related source 9宝玉

2026/9/2 18:25:23View source

Related source 10OpenAI Blog

2026/9/3 00:00:00View 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