MODEL HISTORY

Qwen3.8 2.4T A95B 指标历史

Qwen3.8 2.4T A95B 的公开评测事实历史,共 61 条观测、49 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61仅在事实变化时追加
指标数量49不同量纲保持分离
后续变更12初始观测不计作变更
首次收录2026/08/14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 ↗

精确 API 的“本次未报价”表示该次完整报价未提供这一价格字段,当前值未知,不代表 API 下架或免费;此前报价保留原来源与日期,仅作历史参考。

APPEND-ONLY FACTS

观测记录

按观测时间倒序;同日多项指标分别保留。

Qwen3.8 2.4T A95B 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 v4.3.1Eloelo1628Artificial 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 v4.3.1Eloelo1439Artificial 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 Eloelo1439Artificial 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 v2elo1628Artificial 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 transparencypoints1.0Artificial Analysis ↗第三方评测有效
Artificial Analysis Openness Indexvsource-unversionedOpenness Indexindex27.8Artificial 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Eloelo1628Artificial 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Eloelo1439Artificial Analysis ↗第三方评测有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.25OpenRouter ↗第一方 API 报价有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant缓存写入价格 / 1M Tokenusd_per_1m_tokensUS$2.5OpenRouter ↗第一方 API 报价有效
Artificial Analysis intelligence component evaluationsv4.3AutomationBench-AA partial scorefraction57.2%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 Eloelo1439Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3CritPtfraction20.0%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3GDP.pdf all-passfraction15.6%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 v2elo1628Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3Humanity's Last Examfraction42.4%Artificial Analysis ↗第三方评测有效
Artificial Analysis capability indicesv4.3Intelligence Indexindex40.0Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-LCR v1.1fraction80.3%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscienceindex4.3Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience accuracyfraction31.3%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3AA-Omniscience hallucination ratefraction39.2%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3SciCodefraction54.1%Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.3Terminal-Bench v4.0fraction11.1%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds106.2 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-AAelo1715Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction60.7%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second24.1 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seconds85.5 秒Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.25Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds114.2 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-AAelo1717Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction60.9%Artificial Analysis ↗第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输入价格 / 1M Tokenusd_per_1m_tokensUS$2Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1输出价格 / 1M Tokenusd_per_1m_tokensUS$6Model Coordinate repository snapshots第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second22.4 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seconds91.9 秒Model Coordinate repository snapshots第三方评测有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.2OpenRouter ↗第一方 API 报价有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant缓存读取价格 / 1M Tokenusd_per_1m_tokensUS$0.25OpenRouter ↗第一方 API 报价有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant输入价格 / 1M Tokenusd_per_1m_tokensUS$2OpenRouter ↗第一方 API 报价有效
OpenRouter exact API offering (qwen/qwen3.8-2.4t-a95b)vexact-variant输出价格 / 1M Tokenusd_per_1m_tokensUS$6OpenRouter ↗第一方 API 报价有效
Artificial Analysis capability indicesvpublic-model-v1Agentic Indexindex57.1Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1上下文窗口tokens983,616Model Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1CritPtfraction20.0%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1端到端响应时间seconds52.6 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-AAelo1720Artificial Analysis ↗第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1GDPval-AA normalizedfraction61.0%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 Examfraction42.4%Artificial Analysis第三方评测有效
Artificial Analysis capability indicesvpublic-model-v1Intelligence Indexindex57.7Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-LCRfraction75.3%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1AA-Omniscienceindex4.3Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1生成速度tokens_per_second50.1 Token/sModel Coordinate repository snapshots第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1SciCodefraction51.6%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Tau Bankingfraction49.1%Artificial Analysis第三方评测有效
Artificial Analysis intelligence component evaluationsv4.1Terminal-Bench v2.1fraction82.0%Artificial Analysis第三方评测有效
Artificial Analysis model aggregatevpublic-model-v1首个回答 Token 等待seconds42.7 秒Model Coordinate repository snapshots第三方评测有效