MODEL HISTORY

Claude Opus 5 (Adaptive Reasoning, Max Effort) 指标历史

Claude Opus 5 (Adaptive Reasoning, Max Effort) 的公开评测事实历史,共 21 条观测、21 项指标;只记录来源变化,不插值、不合成本站总分。

数据快照:2026-07-25T00:00:00.000Z查看原始来源 ↗
历史观测21仅在事实变化时追加
指标数量21不同量纲保持分离
后续变更0初始观测不计作变更
首次收录2026/07/25本站可追溯起点
最近观测2026/07/25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。

Claude Opus 5 (Adaptive Reasoning, Max Effort) 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-v1Agentic Indexindex55.3有效
Artificial Analysis model aggregatevpublic-model-v1缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.5有效
Artificial Analysis model aggregatevpublic-model-v1缓存写入价格 / 1M Tokenusd_per_1m_tokensUS$6.25有效
Artificial Analysis capability indicesvpublic-model-v1Coding Indexindex78.0有效
Artificial Analysis model aggregatevpublic-model-v1上下文窗口tokens1,000,000有效
Artificial Analysis intelligence component evaluationsv4.1CritPtfraction29.1%有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds73.6 秒有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AAelo1861有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction68.0%有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction93.2%有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction52.6%有效
Artificial Analysis model aggregatevpublic-model-v1输入价格 / 1M Tokenusd_per_1m_tokensUS$5有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex60.7有效
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction70.0%有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex31.3有效
Artificial Analysis model aggregatevpublic-model-v1输出价格 / 1M Tokenusd_per_1m_tokensUS$25有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second54.8 Token/s有效
Artificial Analysis intelligence component evaluationsv4.1SciCodefraction55.7%有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction30.3%有效
Artificial Analysis intelligence component evaluationsv4.1Terminal-Bench v2.1fraction89.1%有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds64.5 秒有效