MODEL HISTORY

Trinity Large Thinking 指标历史

Trinity Large Thinking 的公开评测事实历史,共 17 条观测、14 项指标;只记录来源变化,不插值、不合成本站总分。

数据快照:2026-07-22T00:00:00.000Z查看原始来源 ↗
历史观测17仅在事实变化时追加
指标数量14不同量纲保持分离
后续变更3初始观测不计作变更
首次收录2026/07/21本站可追溯起点
最近观测2026/07/22不是模型发布日期

怎样理解这份历史

每条记录对应来源快照中的一个原始指标。相同值重复抓取不会制造新记录;只有数值、临时状态或移除状态发生变化时才追加。首个快照是本站的可追溯起点,不代表该指标第一次对外发布。

Index、Elo、百分比等单位不能互相相减,也不会在这里合成为“综合总分”。

APPEND-ONLY FACTS

观测记录

按观测时间倒序;同日多项指标分别保留。

Trinity Large Thinking 的公开评测事实历史
观测日期基准 / 版本指标原始值状态
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex18.2有效
Artificial Analysis capability indicesvpublic-model-v1Coding Indexindex25.8有效
Artificial Analysis capability indicesvpublic-model-v1Agentic Indexindex3.7有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AAelo562有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction3.1%有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction5.6%有效
Artificial Analysis intelligence component evaluationsv4.1Terminal-Bench v2.1fraction20.6%有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex24.5临时 / 估算
Artificial Analysis intelligence component evaluationsv4.1GDPval-AAelo554有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction2.7%有效
Artificial Analysis intelligence component evaluationsv4.1SciCodefraction36.1%有效
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction33.0%有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex-44.2有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction14.7%有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction75.2%有效
Artificial Analysis intelligence component evaluationsv4.1CritPtfraction0.9%有效
Artificial Analysis economics capability indexvunversionedeconomics capability indexindex19.8有效