技术档案

定制AI芯片Custom AI silicon

按客户需求设计芯片

每条都摘自下方编号的来源原文；只写事实，不写预测或建议。

技术简介

  • TPU不同于通用并行处理器，是专门加速大规模AI工作负载的专用集成电路。2
    Unlike general-purpose parallel processors,TPUs are Application-Specific Integrated Circuits (ASICs) built for one purpose: accelerating large-scale AI workloads.

已有应用

  • 谷歌Ironwood TPU已经用于Gemini和Nano Banana模型的训练与推理。2
    Ironwood, our latest generation Tensor Processing Unit (TPU), is the cutting-edge hardware behind advanced models like Gemini and Nano Banana, from massive-scale training to high-throughput, low-latency inference.

主要参与者

  • Marvell与谷歌签署了向谷歌开发定制半导体产品的商业协议。1
    On July 29, 2026, Marvell Technology, Inc. (the “Company”) and Google LLC (“Google”) entered into a commercial agreement relating to the Company’s development of custom semiconductor products to Google (the “Custom Products”).

相关卡片

来源

  1. SEC · 8-K · 2026-08-19
  2. cloud.google.com · Inside the Ironwood TPU codesigned AI stack | Google Cloud Blog · 2025-11-06