MODEL HISTORY

Gemma 4 E4B (Reasoning) 指标历史

Gemma 4 E4B (Reasoning) 的公开评测事实历史,共 65 条观测、37 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65仅在事实变化时追加
指标数量37不同量纲保持分离
后续变更28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。

Gemma 4 E4B (Reasoning) 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 v4.3.1Eloelo177Artificial 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 v2elo177Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedModel availabilitypoints6.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedModel transparencypoints1.0Artificial 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Eloelo177Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3CritPtfraction0.6%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 v2elo177Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3Humanity's Last Examfraction3.8%Artificial Analysis ↗第三方评测有效
Artificial Analysis capability indicesv4.3Intelligence Indexindex8.9Artificial Analysis ↗第三方评测临时 / 估算
Artificial Analysis intelligence component evaluationsv4.3AA-LCR v1.1fraction32.0%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscienceindex-19.7Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience accuracyfraction8.6%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience hallucination ratefraction30.9%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds49.1 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second51.7 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seconds39.5 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seconds52.4 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-AAelo230Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction57.6%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction3.8%Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex12.2Artificial Analysis第三方评测临时 / 估算
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction33.0%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex-19.7Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second48.5 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seconds42.1 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seconds28.7 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-AAelo230Artificial Analysis ↗第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex11.9Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex-19.6Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second89.7 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seconds23.1 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seconds28.1 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-AAelo231Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction52.2%Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex11.9Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second91.7 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seconds22.6 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Agentic Indexindex1.8Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Coding Indexindex9.4Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds27.6 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-AAelo231Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction0.0%Artificial Analysis ↗第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex12.2Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second94.3 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction5.4%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Terminal-Bench v2.1fraction1.9%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds22.3 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1上下文窗口tokens128,000Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1CritPtfraction0.6%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds27.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GPQA Diamondfraction57.6%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Humanity's Last Examfraction3.7%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输入价格 / 1M Tokenusd_per_1m_tokensUS$0.02Model Coordinate repository snapshots第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex12.5Artificial Analysis第三方评测临时 / 估算
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction30.7%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex-20.1Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输出价格 / 1M Tokenusd_per_1m_tokensUS$0.1Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second94.6 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1SciCodefraction24.4%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds22.3 秒Model Coordinate repository snapshots第三方评测有效