GPT-5 · tau2-bench(Airline category) 62.6% · 主文证据(ev-gpt5-tau2-airline)

GPT-5 的 tau2-bench · Airline category 官方出处。GPT-5 (high) · 62.6%。见《Introducing GPT-5 for developers | OpenAI》Detailed benchmarks / τ²-bench airline。证据 ev-gpt5-tau2-airline。

原文与定位

GPT-5 · tau2-bench(Airline category),GPT-5 (high)配置,Category score 62.6%(成绩公布 2025-08-07)。见《Introducing GPT-5 for developers | OpenAI》Detailed benchmarks / τ²-bench airline。 查看成绩条件

Tau 2-bench airline: GPT-5 (high) 62.6%; GPT-5 mini (high) 60.0%; GPT-5 nano (high) 41.0%.
官方材料
Introducing GPT-5 for developers | OpenAI
定位
主文 · Detailed benchmarks / τ²-bench airline
材料版本
Official GPT-5 developer release article; launch-day Wayback replay 2025-08-07 21:34:00 UTC, CDX status 200 text/html; detailed benchmark tables read and transcribed. This is the later of two same-day captures with different FrontierMath nano values.
版本公布
披露时间
· 已有首发依据
采集时间
历史依据
Wayback CDX capture 20250807213400, status 200 text/html; replay via http://web.archive.org/web/20250807213400id_/https://openai.com/index/introducing-gpt-5-for-developers/ returned HTTP 200. The archived article is dated August 7, 2025.
引用文本

GPT-5 · tau2-bench(Airline category),GPT-5 (high)配置,Category score 62.6%(成绩公布 2025-08-07)。官方出处 Introducing GPT-5 for developers | OpenAI(主文,Official GPT-5 developer release article; launch-day Wayback replay 2025-08-07 21:34:00 UTC, CDX status 200 text/html; detailed benchmark tables read and transcribed. This is the later of two same-day captures with different FrontierMath nano values.)Detailed benchmarks / τ²-bench airline,https://openai.com/index/introducing-gpt-5-for-developers/。Benchlog https://benchlog.lekshome.cc/evidence/gpt-5/ev-gpt5-tau2-airline/

关联成绩与条件

GPT-5 · tau2-bench 成绩与配置
指标与成绩配置与评测条件证据关系
Category score62.6%原文 62.6% · 单位 %
公布
GPT-5 (high)
设置
API reasoning model; the developer release benchmark table labels high reasoning effort.
未知项
未另记
工具
允许工具
推理预算
high
框架
未记录
协议
τ²-bench airline
条件
API GPT-5; high reasoning effort.

直接支持该成绩

在排名中查看此成绩