MODEL HISTORY

Kimi K3 (max) 指标历史

Kimi K3 (max) 的公开评测事实历史,共 99 条观测、56 项指标;只记录来源变化,不插值、不合成本站总分。

数据快照:2026-09-18T01:47:51.386Z查看原始来源 ↗
历史观测99仅在事实变化时追加
指标数量56不同量纲保持分离
后续变更43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2026/09/18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。同一口径下相同值重复抓取不会制造新记录;数值、临时状态或来源状态变化时追加,核实后的方法版本另立指标身份。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。当前模型来源为 Artificial Analysis。

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”。

GDPval-AA 与 AA-Briefcase 的实时 Elo 现按裁判协议 v4.3.1 单独记录;旧记录的版本、值与日期保留,未区分裁判的旧值仅作历史参考。Intelligence Index 仍为 v4.3,官方用于指数的 Elo 贡献在模型加入时固定,不能用当前实时 Elo 重算指数。核对方法说明 ↗

APPEND-ONLY FACTS

观测记录

按观测时间倒序;同日多项指标分别保留。

Kimi K3 (max) 的公开评测事实历史
观测日期基准 / 版本指标原始值来源状态
Artificial Analysis GDPval-AA任务 v2 · 裁判 v4.3.1实时 Elo · 裁判协议 v4.3.1Eloelo1548Artificial Analysis ↗第三方评测有效
Artificial Analysis AA-Briefcase来源未标任务版本 · 裁判 v4.3.1实时 Elo · 裁判协议 v4.3.1Eloelo1488Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3.1实时 Elo · 裁判协议 v4.3.1AA-Briefcase overall Eloelo1488Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3.1实时 Elo · 裁判协议 v4.3.1GDPval-AA v2elo1548Artificial Analysis ↗第三方评测有效
Artificial Analysis AA-Briefcasevsource-unversioned历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1488Artificial Analysis ↗第三方评测有效
Artificial Analysis GDPval-AAv2历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1552Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较AA-Briefcase overall Eloelo1488Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AA v2elo1552Artificial Analysis ↗第三方评测有效
Artificial Analysis GDPval-AAv2历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1566Artificial Analysis ↗第三方评测有效
Artificial Analysis AA-Briefcasevsource-unversioned历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1492Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较AA-Briefcase overall Eloelo1492Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AA v2elo1566Artificial Analysis ↗第三方评测有效
Artificial Analysis GDPval-AAv2历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1569Artificial Analysis ↗第三方评测有效
Artificial Analysis AA-Briefcasevsource-unversioned历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1493Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较AA-Briefcase overall Eloelo1493Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AA v2elo1569Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedModel availabilitypoints4.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedModel transparencypoints3.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedOpenness Indexindex38.9Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedPosttraining data accesspoints0.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedPosttraining data licensepoints0.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedPretraining data accesspoints0.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedPretraining data licensepoints0.0Artificial Analysis ↗第三方评测有效
Artificial Analysis GDPval-AAv2历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1572Artificial Analysis ↗第三方评测有效
Artificial Analysis AA-Briefcasevsource-unversioned历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较Eloelo1495Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较AA-Briefcase overall Eloelo1495Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AA v2elo1572Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AutomationBench-AA partial scorefraction58.3%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较AA-Briefcase overall Eloelo1497Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3CritPtfraction23.4%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3GDP.pdf all-passfraction22.0%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AA v2elo1584Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3Humanity's Last Examfraction46.9%Artificial Analysis ↗第三方评测有效
Artificial Analysis capability indicesv4.3Intelligence Indexindex43.8Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-LCR v1.1fraction88.7%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscienceindex19.7Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience accuracyfraction47.6%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience hallucination ratefraction53.2%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3SciCodefraction59.5%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3Terminal-Bench v4.0fraction12.6%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds79.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AAelo1670Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction58.5%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second35.6 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds65.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Agentic Indexindex54.3Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds70.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AAelo1675Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction58.8%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction46.9%Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex59.7Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction82.7%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex19.7Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second39.0 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction46.0%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds57.6 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis Analyst Agentv80-tasks-5-attemptspass^5percent38.8%Artificial Analysis ↗第三方评测有效
Artificial Analysis Analyst Agentv80-tasks-5-attemptspass@1percent57.8%Artificial Analysis ↗第三方评测有效
Artificial Analysis Analyst Agentv80-tasks-5-attemptspass@5percent72.5%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds219.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AAelo1687Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction59.4%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second33.3 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds204.4 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds239.1 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AAelo1686Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction59.3%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second32.1 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds223.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds73.9 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second36.1 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds60.1 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Agentic Indexindex50.1Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.3Model Coordinate repository snapshots第三方评测有效
Artificial Analysis agentic capability indexvunversionedagentic capability indexindex50.1Artificial Analysis第三方评测有效
Artificial Analysis coding capability indexvunversionedcoding capability indexindex76.2Artificial Analysis第三方评测有效
Artificial Analysis finance capability indexvunversionedfinance capability indexindex48.3Artificial Analysis第三方评测有效
Artificial Analysis legal capability indexvunversionedlegal capability indexindex50.0Artificial Analysis第三方评测有效
Artificial Analysis healthcare capability indexvunversionedhealthcare capability indexindex42.5Artificial Analysis第三方评测有效
Artificial Analysis engineering capability indexvunversionedengineering capability indexindex57.8Artificial Analysis第三方评测有效
Artificial Analysis economics capability indexvunversionedeconomics capability indexindex55.7Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Coding Indexindex76.2Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1上下文窗口tokens1,048,576Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1CritPtfraction23.4%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds67.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1历史参考 · 原记录未区分当前裁判协议,不作跨协议能力变化比较GDPval-AAelo1679Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction59.0%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction93.5%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction44.3%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输入价格 / 1M Tokenusd_per_1m_tokensUS$3Model Coordinate repository snapshots第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex57.1Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction74.7%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex18.4Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输出价格 / 1M Tokenusd_per_1m_tokensUS$15Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second39.5 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1SciCodefraction58.7%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction33.4%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Terminal-Bench v2.1fraction85.0%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds54.9 秒Model Coordinate repository snapshots第三方评测有效