2026年10月27日 (英语).
人工分析情报索引v5
Index v5具有2个新的基准测试和3个升级功能。 指数的绝大部分首次建立在私人数据集之上
人工分析情报指数是我们衡量各种任务模型能力的综合尺度，反映当今现实世界的使用情况。 v5是我们自一月份v4 发布以来最大的升级。 有了v5，指数的60%现在基于有私人测试集的评价或被搁置的答案。
除了情报指数v5，我们发布人工分析进度指数v1
新到情报索引：
FrontierCode 2.0:Cognition的新-发布私人的基准测试真实世界编码任务，分级于一个寄存器维护者是否会合并agent的更改
终端-奔驰-科学：70项代理科学研究任务，在终端环境中运行
升级：
HLE-Diamond (取代"人类最后的考试")：精细的"人类最后的考试"1000个问题版本，在推理和知识问题之间平分。
AA-Omniscience v1.1：更新数据集，删除了饱和和和公开的问题，增加了新的问题，并修订了模棱两可或错误的问题
CritPt v2：由 CritPt 作者修改并使用代码执行工具运行的 CritPt v2 的一个私密的40个问题硬子集
留在索引中：AA-Briefcase，自动化 Bench-AA,GDP.pdf和终端-Bench 4.0
从指数中删除：GDPval-AA v2.1,SciCode和AA-LCR v1.1
智能指数 v5 权重：
知识工作(25%):AA-Briefcase(15%)、自动化本克-AA(5%)和GDP.pdf(5%)
编码(25%)：边疆代码2.0(15%)和终点站-奔跑4.0(10%)
科学(25%)：终端-Bench-Science(15%)和CritPt v2硬子集(10%)
一般(25%):AA-Omniscience v1.1 15%(准确性10%，无卤化5%)和HLE-Diamond(10%)
主要成果：
Claude Opus 5.5 (最大)领先于58分，领先于GPT-6 Astra (最大，57分)和GPT-6.1 Sol (最大，55分):Anthropic和OpenAI持有前5个型号。 Gemini 4 Argon (高) 是下一个在50
开放权重差距从3分扩大到12分：Kimi K3.1 Pro(重量尚未达到发布)是46分的最高开放权重型号。 在发布重量中，DeepSeek V4.1闪光1017(最大)和GLM-5.3(最大)领先于领先者后32-26分并低于Claude Haiku 5.5分(最大，36分).
每个任务的成本在主要模型中相差超过13x：在显示的10个得分最高的模型中，GPT-6.1 Terra(最大)是每个情报指数任务0.96美元最便宜的，而Claude有线5.1(最大)是13.04美元最昂贵的。
FrontierCode 2.0是Cognition私人编码基准测试的最新更新。 在今天早些时候，发布包含132项任务在真实的开源寄存器中，分级为维护者是否会将agent的改变合并。 尝试失败任何屏蔽检查得分0，任务保持隐私以避免被污染。 我们在斯特鲁普三度执行每一项任务我们的特长
Claude Opus 5.5 (最大)领先于46%，其次是GPT-6 Astra (最大),GPT-6.1 Terra (最大)和Claude Sonnet 5.5 (最大)，均占42%. DeepSeek V4.1 Flash 1017 (max)是最高得分的开放权重型号，占28%，落后于领跑者18分。
终端-Bench-Science测试agents能否在终端进行科学研究工作流程。 专家负责和审查每项任务，一项任务只有在每次试验都通过时才能通过。
GPT-6 Astra (最大)以63%领先，其次是Claude Opus 5.5 (最大)以59%领先，GPT-6.1 Sol (最大)以58%领先。 得分大大低于领先者：最佳开放权重型，Kimi K3.1 Pro(重量还不到发布)，得分34%，而Claude Haiku 5.5 (最大)得分20%. 最佳开放权重模型有发布重量，DeepSeek V4.1 Flash 1017 (最大)，得分为13%，大多数开放权重模型我们的基准分低于10%.
HLE-Diamond取而代之的是"索引"中"人类最后的考试". 这是来自AI安全中心(CAIS)和"规模AI"的精细的1000个问题版本"人类最后考试"，在推理和知识之间平分。 Claude Opus 5.5 (最大)以68%领先，其次是GPT-6 Astra (最大)以64%领先，Claude 有线5.1 (最大)以60%领先。
知识问题比主要模型的推理问题更难：Claude Opus 5.5(最大)和GPT-6 Astra(最大)在推理上都得分78%，而知识上得分分别为58%和50%.
AA-Omniscience v1.1 是我们知识与幻觉基准测试的清版，得分大多与v1一致，模型在AA-Omniscience v1.1指数(中位0.6)上平均得分较低。 发布是一个新的公共子集，在Hugging Face, AA-Omniscience-Public-v1.1上
Index对一个私人的40个问题硬子集CritPt v2的分数：70个研究级物理问题，减去发布所有5个尝试中解决的7个前沿模型的30个。 模特儿在Stirrup的Python外线沙盒里工作我们的特长带
GPT-6 Astra以72%的高功率领先，其次是GPT-6.1 Sol以69%的最高功率领先。 Claude Opus 5.5 (最大)，指数整体中最高的模型，得分为61%. Kimi K3.1 号机车 Pro(重量尚未达到发布)是最高得分的开放权重型号，为45%.
v5为所有9个评价都留下了很大的进展空间，没有一个单一模式领导所有评价。 在我们测量的模型中，Claude Opus 5.5领先于3个，而其他5个模型领先于其余6个：
Claude Opus 5.5: FrontCode 2.0 (46%),HLE-Diamond (68%)和AA-Omniscience v1.1 (48)
GPT-6 Astra：终端-Bench-Science(63%)和CritPt v2硬子集(72%)
Claude Sonnet 5.5: AA-Briefcase (1812 Elo) (英语).
Kimi K3.1 Pro：终点站-奔驰 4.0 (67%)
Gemini 4 边形：自动化 Bench-AA(78%)
Muse Wonder Full: GDP.pdf (33%全通) (中文(简体) ).
开放权重模型在v5中进一步落后。 最佳成绩为Kimi K3.1 Pro(积分尚未达到发布)，得分46分，落后于领跑者12分，相较于v4.3相差3分。 在有发布重量的模型中，DeepSeek V4.1 Flash 1017 (max)和GLM-5.3 (max)的得分最高，分别以32分和Claude Haiku 5.5 (max)的得分超过全部。
在HLE-Diamond和Terminal-Bench-Science上差距最大，其中Kimi K3.1 Pro的得分为37%和34%，而Claude Opus 5.5(最大)为68%，而GPT-6 Astra(最大)为63%.
主要模式的成本差异很大。 Claude Opus 5.5(最大)每次情报指数任务费用为10.04美元，GPT-6 Astra(最大)为6.24美元，GPT-6.1 Sol(最大)为1.27. GPT-6.1 Sol(最大)在Claude Opus 5.5(最大)的4分以每任务成本的8分左右。
努力设置在v5中比在v4.3中更重要。 Claude Opus 5.5分分35分，低功率分58分，最高分为10分，按每个情报指数任务的成本(0.98美元对10.04美元)计算。 Claude Sonnet 5.5从低到最大跨出29个分数，从v4.3的20个起起，而GPT-6.1 Sol的分布从10个分数到20个分数。
v5任务要求模型进行大量的工作。 我们测得的每个模型，每个情报指数任务产生的输出信使都比v4.3多，中位数上升了三分之二，从35克上升到59克。
代理评价占大多数象征性用途：终端-Bench-Science、终端-Bench 4.0、AA-Briefcase和FrontierCode 2.0占指数重量的55%，但产出符号的中位数~90%
Terminal-Bench-Science 是最象征性的密集型：它占每个任务产出的大约三分之一
除了v5，我们是发布人工分析进度指数v1. 它将自2022年11月以来的语言型号发布置于一个能力规模上，跨越了广阔的基准测试。
使用进步指数来比较基础能力如何随时间变化，包括在相隔发布年的模型之间。 为了比较当前使用案例的模型并了解目前的前沿，使用情报指数，这仍然是我们目前模型情报的主要衡量标准。
感谢Epoch AI和Anthropic——我们在设计进步指数时借鉴了Epoch能力指数的学习。
探索进步指数：人为分析.ai/趋势
方法：人工分析.ai/方法/进度指数
感谢Cognition，史蒂文·迪尔曼和终端-奔驰团队，AI安全中心，规模AI，以及敏辉朱，敏阳天，郝鹏和克里特普特团队为本次更新背后的基准测试出场。
探索全部结果：人工分析.ai
方法：人工分析.ai/方法/情报基准
读取最新的
Code2Video-AA:AI模型如何在代码中设计视频？.
人工分析"HeyGen's Code2Video Bench"的应用，评价模型在使用HyperFrames HTML的114个设计简报中创建了多好的产品和品牌视频。
2026年10月26日 (英语).
GPT-6.1 Terra推动成本效益前沿
OpenAI拥有发布 GPT-6.1 Terra，推动成本效率前沿。 这是自GPT-5.6以来的第一个Terra级模型，其智能性小于成本的三分之一。
2026年10月26日 (英语).
Kimi K3.1 号机车发布在情报指数第4位，并将领导开放权重模型
Kimi K3.1 号机车 Pro 发布位于人工分析智能指数第4位，并将领导开放权重模型
2026年10月23日 (英语).
本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
October 27, 2026
Announcing Artificial Analysis Intelligence Index v5
Index v5 features 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks reflecting real-world usage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: an updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Intelligence Index v5 weights:
➤ Knowledge Work (25%): AA-Briefcase (15%), AutomationBench-AA (5%), and GDP.pdf (5%)
➤ Coding (25%): FrontierCode 2.0 (15%) and Terminal-Bench 4.0 (10%)
➤ Science (25%): Terminal-Bench-Science (15%) and CritPt v2 hard subset (10%)
➤ General (25%): AA-Omniscience v1.1 15% (10% accuracy, 5% non-hallucination) and HLE-Diamond (10%)
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
FrontierCode 2.0 is the latest update of Cognition's private coding benchmark. Launched earlier today, it comprises 132 tasks in real open-source repositories, graded on whether a maintainer would merge the agent's change. An attempt that fails any blocking check scores 0, and the tasks stay private to avoid contamination. We run every task three times in Stirrup, our agentic harness.
Claude Opus 5.5 (max) leads at 46%, followed by GPT-6 Astra (max), GPT-6.1 Terra (max) and Claude Sonnet 5.5 (max), all at 42%. DeepSeek V4.1 Flash 1017 (max) is the highest-scoring open weights model at 28%, 18 points behind the leader.
Terminal-Bench-Science tests whether agents can carry out scientific research workflows in a terminal. Experts curate and review each task, and a task passes only if every one of its tests passes.
GPT-6 Astra (max) leads at 63%, followed by Claude Opus 5.5 (max) at 59% and GPT-6.1 Sol (max) at 58%. Scores drop off sharply below the leaders: the best open weights model, Kimi K3.1 Pro (weights not yet released), scores 34%, and Claude Haiku 5.5 (max) scores 20%. The best open weights model with released weights, DeepSeek V4.1 Flash 1017 (max), scores 13%, and most open weights models we benchmarked score under 10%.
HLE-Diamond replaces Humanity's Last Exam in the Index. It is the refined 1,000-question version of Humanity's Last Exam from the Center for AI Safety (CAIS) and Scale AI, split evenly between reasoning and knowledge. Claude Opus 5.5 (max) leads at 68%, followed by GPT-6 Astra (max) at 64% and Claude Fable 5.1 (max) at 60%.
Knowledge questions are harder than reasoning questions for the leading models: Claude Opus 5.5 (max) and GPT-6 Astra (max) both score 78% on reasoning, compared with 58% and 50% on knowledge.
AA-Omniscience v1.1 is a cleaned version of our knowledge and hallucination benchmark, with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised. Scores are mostly aligned with v1, with models scoring on average 2 points lower on the AA-Omniscience v1.1 Index (median 0.6). We’ve also released a new public subset on Hugging Face, AA-Omniscience-Public-v1.1, recalibrated to ensure accuracy, hallucination rate, and overall performance are consistent between the public and private sets.
The Index scores CritPt v2 on a private 40-question hard subset: the 70 research-level physics questions, minus the 30 that seven frontier models solved in all five attempts at launch. Models work in an offline Python sandbox in Stirrup, our agentic harness.
GPT-6 Astra leads at 72% on high effort, followed by GPT-6.1 Sol at 69% on max effort. Claude Opus 5.5 (max), the top model in the Index overall, scores 61%. Kimi K3.1 Pro (weights not yet released) is the highest-scoring open weights model at 45%.
v5 leaves significant room for progress across all nine evaluations, and no single model leads them all. Of the models we measured, Claude Opus 5.5 leads three, and five other models lead the remaining six:
➤ Claude Opus 5.5: FrontierCode 2.0 (46%), HLE-Diamond (68%) and AA-Omniscience v1.1 (48)
➤ GPT-6 Astra: Terminal-Bench-Science (63%) and CritPt v2 hard subset (72%)
➤ Claude Sonnet 5.5: AA-Briefcase (1812 Elo)
➤ Kimi K3.1 Pro: Terminal-Bench 4.0 (67%)
➤ Gemini 4 Argon: AutomationBench-AA (78%)
➤ Muse Wonder Full: GDP.pdf (33% all-pass)
Open weights models fall further behind in v5. The best, Kimi K3.1 Pro (weights not yet released), scores 46, 12 points behind the leader, compared with a 3-point gap in v4.3. Of models with released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) score highest at 32, and Claude Haiku 5.5 (max) outscores all of them.
The gap is widest on HLE-Diamond and Terminal-Bench-Science, where Kimi K3.1 Pro scores 37% and 34%, compared with 68% for Claude Opus 5.5 (max) and 63% for GPT-6 Astra (max).
Cost varies widely among the leading models. Claude Opus 5.5 (max) costs $10.04 per Intelligence Index task, GPT-6 Astra (max) $6.24 and GPT-6.1 Sol (max) $1.27. GPT-6.1 Sol (max) scores within 4 points of Claude Opus 5.5 (max) at about an eighth of the cost per task.
Effort settings matter more in v5 than in v4.3. Claude Opus 5.5 scores 35 at low effort and 58 at max, at 10x the cost per Intelligence Index task ($0.98 vs $10.04). Claude Sonnet 5.5 spans 29 points from low to max, up from 20 in v4.3, and GPT-6.1 Sol's spread doubles from 10 to 20 points.
v5 tasks demand substantially more work from models. Every model we measured generates more output tokens per Intelligence Index task than in v4.3, with the median rising by two thirds, from 35k to 59k.
➤ Agentic evaluations account for most token use: Terminal-Bench-Science, Terminal-Bench 4.0, AA-Briefcase and FrontierCode 2.0 make up 55% of the Index weight but a median ~90% of output tokens
➤ Terminal-Bench-Science is the most token-intensive: it accounts for about a third of output Tokens per Task on its own
Alongside v5, we're launching the Artificial Analysis Progress Index v1. It places language models released since November 2022 on one capability scale, across a wide span of benchmarks.
Use the Progress Index to compare how underlying capability has changed over time, including between models released years apart. To compare models for current use cases and understand the present frontier, use the Intelligence Index, which remains our primary measure of current model intelligence.
Thank you to Epoch AI and Anthropic - we built on learnings from the Epoch Capabilities Index in designing the Progress Index.
Explore the Progress Index: artificialanalysis.ai/trends
Methodology: artificialanalysis.ai/methodology/progress-index
Thank you to Cognition, Steven Dillmann and the Terminal-Bench team, the Center for AI Safety, Scale AI, and Minhui Zhu, Minyang Tian, Hao Peng and the CritPt team for the benchmarks behind this update.
Explore the full results: artificialanalysis.ai
Methodology: artificialanalysis.ai/methodology/intelligence-benchmarking
Read the latest
Code2Video-AA: how well do AI models design videos in code?
Artificial Analysis' implementation of HeyGen's Code2Video Bench, evaluating how well models create short product and brand videos from 114 design briefs using HyperFrames HTML.
October 26, 2026
GPT-6.1 Terra pushes the cost efficiency frontier
OpenAI has released GPT-6.1 Terra, pushing the cost efficiency frontier. It is the first Terra-class model since GPT-5.6, and is more intelligent for less than a third of the cost
October 26, 2026
Kimi K3.1 Pro launches at 4th on the Intelligence Index and will lead open weights models
Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index and will lead open weights models
October 23, 2026
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.