Google Launches Two In-House Chips: TPU 8TX and TPU 8I Max
English text is being prepared.
Google announced the launch of two new in-house tensor processors, TPU 8TX and TPU 8I Max, expanding the eighth-generation TPU family to four systems within a year. TPU 8TX focuses on full-stack confidential inference; it is the first TPU system to integrate hardware confidential computing features, and it introduces the Boardfly network topology, reducing the inter-chip communication diameter by more than 50% and halving inference latency compared with the previous generation. TPU 8I Max uses a dual-chip package, integrating two 8I chips within a single module, providing 17.2 TB/s of memory bandwidth, 768 MB of on-chip SRAM and 384GB of HBM; within the same rack space and power consumption, its compute density and data throughput are twice those of TPU 8I. Both chips and the acceleration software will be available on Google Cloud.
Why read this
AI score 62
The original text presents the positioning and key specifications of the two TPUs, allowing readers to understand the division of labor of Google's in-house compute in confidential inference and large-model inference.
Topics
Body
Google昨（9）日宣布推出两款新自研张量处理器（TPU）「TPU 8TX」与「TPU 8I Max」，全面扩大抢食AI商机。
业界分析，Google第八代TPU家族从尖端制程到系统级封装，目前绑定台积电生态系，随着Google持续扩大相关产品线阵容，并获得更多大厂采用，台积电跟着吃补。
Google Cloud AI与算力基础架构副总裁暨总经理Mark Lohmeyer表示，Google过去每两年或一年推出一款AI芯片的周期将加速，现在已达到「单年度推出四款专用系统」的境界，以极速迭代的硬件研发能量，因应市场需求。
Google上半年才推出针对超大规模训练的TPU 8T与着重低延迟推理的TPU 8I，如今再度拓展第八代TPU家族，推出TPU 8TX与TPU 8I Max，达成一年内发表四款TPU系统的里程碑。
业界分析，AI技术发展重心由「纪录系统（Systems of Record）」跨入能与企业ERP、数据库联动并自主采取行动的「行动系统（Systems of Action）」，底层算力负载倍数暴增。
Google两款新芯片各具战略定位。TPU 8TX主打全链路机密推理，是首款整合硬件机密计算特性的TPU系统，与负责系统控制的AMD x86处理器架构深度整合。芯片通过端到端PCIe链路硬件加密与身份认证，确保金融、医疗或严格监控行业在推理过程中，模型权重与私有数据均处于隔离受保护状态；同时导入全新网络拓扑架构Boardfly，将芯片间的通讯直径缩小超过50%，让推理延迟较前代大幅减半。
TPU 8I Max主打承载巨量文档、代码库与并行代理流程设计，采用双芯片封装架构，于单一模块内整合两颗8I芯片，提供高达17.2 TB／s的存储器带宽（提升二倍）、768 MB片上SRAM（提升二倍）与384GB的HBM（增加33%），在相同的机架空间与能耗条件下，带来优于TPU 8I 两倍的算力密度与数据吞吐能力。
Google强调，未来的算力架构已非单纯比拼加速器颗数，而是硅芯片、网络互联、存储器带宽与调度软件的全方位整合设计。新芯片与加速软件将全面于Google Cloud上线，为全球企业构筑兼具顶级性能、弹性调度与高合规标准的次世代AI基础设施。