智通财经APP获悉，微软(MSFT.US)于当地时间周一正式推出全新研发的Mu小尺寸语言模型，这款专为个人电脑端本地化运行设计的AI工具引发市场关注。根据微软官方技术文档披露，该模型将率先应用于Windows系统"设置"功能的智能代理模块，通过自然语言处理技术将用户输入转化为系统指令，目前已在搭载Copilot+功能的开发者预览版Windows系统中向Insider计划参与者开放测试。
值得关注的是，微软特别强调Mu模型实现了硬件层面的深度优化，所有计算任务完全卸载至神经处理单元NPU执行，实测响应速度突破每秒100个token的处理阈值，充分满足系统设置场景下对即时交互体验的严苛要求。在模型训练环节，微软采用英伟达(NVDA.US)A100数据中心GPU构建训练集群，依托Azure机器学习平台完成全流程算法优化。
为确保终端设备性能最大化，微软与AMD(AMD.US)、英特尔(INTC.US)、高通(QCOM.US)三家芯片厂商展开深度协作，针对不同架构的NPU单元定制开发量化运算方案。这种跨产业协同策略既保障了模型在各类终端设备的兼容性，又通过硬件加速技术实现了能效比的显著提升。
受新品发布消息提振，微软股价在当日交易时段维持稳中有升态势，截至发稿，微软夜盘微涨0.44%，报488.13美元/股，资本市场对这家科技巨头在端侧AI领域的布局给予积极回应。
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.