英特尔本季财报并不差，出问题的是下一季：营收和每股收益指引大幅低于相对一致预期，① PC 被内存按住，②以色列工厂又出了事，南部基亚特加特的Fab 28，是公司早年 10纳米量产主力之一，至今仍扛着一批服务器用10纳米产品，本月初爆炸停产约两周。有分析师再问明年资本开支多少、自由现金流何时转正？首席财务官戴夫·辛斯纳（Dave Zinsner）的回答是：“大概不会有正的自由现金流”。CEO 陈立武: 供应卡点不止晶圆厂，管理层提到，Intel 3 产出在爬，但行业里先进有机基板紧缺，EMIB #先进封装 跟着受限——晶圆做出来了，封装材料跟不上，服务器 CPU 仍交不出去。

How Battery Management Systems Are Improving Automotive Battery Performance Introduction Modern automotive batteries rely on electronic monitoring and control systems to operate safely and efficiently. A battery management system, commonly known as a BMS, monitors battery conditions and helps ensure that cells operate within specified limits. Its role becomes especially important in electric vehicles, where large battery packs contain many interconnected cells. As vehicle electrification progresses, battery management is becoming an important source of innovation in the automotive batteries market. Monitoring Battery Conditions A battery management system monitors parameters such as voltage, current, and temperature. These measurements help estimate the battery's state of charge and state of health. Accurate monitoring allows vehicle control systems to make informed decisions about available energy, charging limits, and power delivery. Reliable estimates also help drivers understand remaining range and battery condition. Improving Battery Safety Battery packs must operate within defined electrical and thermal limits. A BMS can detect abnormal conditions and initiate protective responses, such as reducing current or disconnecting the battery through appropriate safety hardware. Monitoring cannot eliminate every battery risk, so manufacturers also use thermal management, mechanical protection, electrical isolation, and validated safety procedures. Supporting Battery Life Charging patterns, temperature exposure, and operating conditions influence battery degradation. A BMS can help reduce unnecessary stress by controlling charging and discharging within approved limits. Cell balancing is another important function. It helps manage differences between individual cells so that variations in charge do not unnecessarily restrict the usable capacity of the pack. Software and Predictive Maintenance Advances in software and data analysis are expanding the capabilities of battery management systems. Depending on the vehicle architecture, algorithms may identify unusual performance patterns, improve battery-state estimates, or support predictive maintenance. Connected vehicle systems can also provide information that helps service teams assess battery condition and plan maintenance. Conclusion Battery management systems are essential to the reliability of modern automotive batteries. Through monitoring, protection, cell balancing, and improved battery-state estimation, they support safer operation and more effective energy utilization. Continued software and hardware innovation will remain important as battery packs become more sophisticated. Source; https://researchintelo.com/report/automotive-batteries-market

#英特尔 在Hot Chips 2026上正式展示下一代服务器处理器Diamond Rapids，以最高256个P核心和全新Fan-out Fabric架构为核心卖点，将产品定位于企业级AI Agent基础设施，直面AMD和Arm日益激烈的竞争。Diamond Rapids将采用英特尔18A-P制程打造，支持高达1.6TB/s内存带宽、128条PCIe Gen 6通道及CXL 3.0，较上代Xeon 6核心数提升50%以上。从物理构造来看，整颗芯片由22个独立裸片拼接而成，具体包括16个基于Intel 18A-P制程的核心小芯片（Core Chiplet）、2个基于Intel 3制程的Fabric Hub以及4个基于Intel 3-T制程的基础裸片（Base Tile），各部分通过Foveros Direct 3D混合键合及UCIe-S铜互联接口相连。这一设计思路与AMD EPYC Venice采用多CPU小芯片围绕大型I/O裸片的方案高度相似。英特尔将在 #AIDC 数据中心CPU市场与英伟达 Vera及AMD Venice展开正面竞争。

#苹果 正针对中国市场单独训练专属大语言模型，并在阿里巴巴的支持下完成开发，有望成为首家获批在华推出自有AI模型的外国企业。此外，Apple Intelligence预计数月内落地中国，采用“双轨并行”策略，除自有模型外，亦将整合阿里巴巴通义千问及百度等本土AI技术。

#谷歌 推出安卓版Gemini Intelligence，将AI深度融入系统，实现跨App自动执行任务等智能操作，被称“安卓版豆包手机”。同时首发专为Gemini设计的Googlebook笔记本，全新智能光标让用户指向内容即可下达指令。安卓生态全面AI化，与苹果即将亮相的Apple Intelligence正面交锋。#AI外设

SemiAnalysis近日称，AI工作负载的范式正从简单的文本生成向复杂的智能体和强化学习演进，CPU正面临"极其严重的产能短缺"。市场研究机构TrendForce的最新报告印证了这一判断：当前AI数据中心的CPU与GPU配比约为1:4至1:8，而在智能体AI时代，这一比例预计将演变至1:1至1:2。这一结构性转变已在供需两端引发连锁反应。#Intel 和 #AMD 已于2026年第一季度末对部分 #CPU 产品线提价。与此同时，英伟达和Arm双双于2026年3月宣布进军服务器CPU市场。

#小米 发布会：新SU7 21.99万起全系下沉激光雷达，首款手机龙虾正式封测。雷军强调，未来三年，小米计划在AI领域投入的资金将超过600亿元，力求在这一前沿赛道取得领先地位。一度被全网猜测为DeepSeek V4的“神秘模型”，最终被小米“认领”。近日一款名为Hunter Alpha的AI模型在开发者平台OpenRouter匿名上线，具备逾1万亿参数规模及100万token上下文窗口，发布后总调用量迅速登顶。因参数规格与外界流传的DeepSeek V4信息高度吻合，一度引发大规模猜测，就连龙虾之父Peter Steinberger也在X上发帖询问，有没有人知道模型的具体细节。MiMo-V2-Pro是此次发布的核心。该模型拥有逾1万亿总参数、420亿活跃参数及100万token上下文窗口，在全球大模型综合智能排行榜Artificial Analysis Intelligence Index上位列全球第八、中国模型第二，超越xAI Grok，仅次于Gemini 3.1 Pro Preview、GPT-5.4、GPT-5.3 Codex、Claude Opus 4.6、Claude Sonnet 4.6及GLM-5。

英特尔股价周一一度上涨3.3%，公司宣布其Xeon 6处理器将作为主CPU应用于英伟达DGX Rubin NVL8系统。机构认为，这标志着 #INTEL 英特尔与英伟达的合作从概念走向落地，未来可围绕产品发布时间与潜在营收空间进行跟踪，AI服务器领域的协同效应开始显现。

#INTEL 英特尔CEO Lip-Bu Tan周二在Cisco AI Summit上表示，公司计划制造图形处理器GPU，并称已聘请高通资深高管Eric Demmers担任GPU首席架构师。

#INTEL Q4营收136.7亿美元，同比下降4.1%；非GAAP口径下调整后每股收益（EPS）为0.15美元，同比增长15.4%调整后毛利率为37.9%客户计算业务（CCG）四季度营收81.9亿美元，同比下降6.6%数据中心和人工智能（DCAI）四季度营收47.4亿美元，同比增长8.9%代工业务（Intel Foundry）四季度营收45.1亿美元，同比增长3.8%18A是本次财报中被反复强调的关键词。英特尔宣布，该工艺已在亚利桑那与俄勒冈实现高产量制造，成为首个在美国本土量产的领先制程节点。从战略层面看，这一进展意味着英特尔已从“路线图承诺”走向“实质交付”，也解释了市场在财报前对其制程能力重新定价的原因。但从财务角度看，18A仍处于投入期：良率爬坡、初期成本与产能限制，继续对毛利率形成压力。