GPT-6 Astra · ARC-AGI-3(Responses API harness with two settings changed to better match real-world performance) 99.9% · 主文证据(ev-astra-0038)

GPT-6 Astra 的 ARC-AGI-3 · Responses API harness with two settings changed to better match real-world performance 官方出处。GPT-6 Astra table result · 99.9%。见《GPT-6 Astra: A new generation of intelligence | OpenAI》Main article / Table 8 / row “ARC-AGI-3” / column “GPT-6 Astra”; Footnote 1。证据 ev-astra-0038。

原文与定位

GPT-6 Astra · ARC-AGI-3(Responses API harness with two settings changed to better match real-world performance),GPT-6 Astra table result配置,Score 99.9%(具体成绩公布时间未确认)。见《GPT-6 Astra: A new generation of intelligence | OpenAI》Main article / Table 8 / row “ARC-AGI-3” / column “GPT-6 Astra”; Footnote 1。 查看成绩条件

ARC-AGI-3 | GPT-6 Astra: 99.9% 1 Footnote 1: On ARC-AGI-3, GPT-6 Astra was run with our responses API harness⁠, which changes two settings to better match real-world performance. The changes do not specifically target ARC-AGI-3.
官方材料
GPT-6 Astra: A new generation of intelligence | OpenAI
定位
主文 · Main article / Table 8 / row “ARC-AGI-3” / column “GPT-6 Astra”; Footnote 1
材料版本
官方英文主文全文及可见表格;原始发布日 2026-09-03;当前页面修订时间未知
版本公布
未确认
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
暂无可靠历史版本依据
引用文本

GPT-6 Astra · ARC-AGI-3(Responses API harness with two settings changed to better match real-world performance),GPT-6 Astra table result配置,Score 99.9%(具体成绩公布时间未确认)。官方出处 GPT-6 Astra: A new generation of intelligence | OpenAI(主文,官方英文主文全文及可见表格;原始发布日 2026-09-03;当前页面修订时间未知)Main article / Table 8 / row “ARC-AGI-3” / column “GPT-6 Astra”; Footnote 1,https://openai.com/index/gpt-6-astra/。Benchlog https://benchlog.lekshome.cc/evidence/gpt-6-astra/ev-astra-0038/

关联成绩与条件

GPT-6 Astra · ARC-AGI-3 成绩与配置
指标与成绩配置与评测条件证据关系
Score99.9%原文 99.9% · 单位 %
公布 未确认
GPT-6 Astra table result
设置
The article reports the maximum score across efforts; exact effort is not specified for this row.
未知项
Do not infer Max or another specific effort from the table alone.
工具
工具条件未记录
推理预算
未记录
框架
responses API harness
协议
未说明
条件
GPT-6 Astra; table reports the maximum score across efforts; the specific effort is not disclosed unless an exact chart result identifies it. Responses API harness; two settings changed to better match real-world performance. ARC-AGI-3 使用 Responses API harness,两项设置调整用于贴近真实表现,未专门针对 ARC-AGI-3。

直接支持该成绩

在排名中查看此成绩