<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>大模型部署 on 知识铺的博客</title>
    <link>https://index.zshipu.com/ai002/tags/%E5%A4%A7%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/</link>
    <description>Recent content in 大模型部署 on 知识铺的博客</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh-CN</language>
    <lastBuildDate>Thu, 03 Sep 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://index.zshipu.com/ai002/tags/%E5%A4%A7%E6%A8%A1%E5%9E%8B%E9%83%A8%E7%BD%B2/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>32GB显存如何跑通56GB大模型：Shared Memory到异构内存架构解析</title>
      <link>https://index.zshipu.com/ai002/post/20260903/32GB%E6%98%BE%E5%AD%98%E5%A6%82%E4%BD%95%E8%B7%91%E9%80%9A56GB%E5%A4%A7%E6%A8%A1%E5%9E%8BShared-Memory%E5%88%B0%E5%BC%82%E6%9E%84%E5%86%85%E5%AD%98%E6%9E%B6%E6%9E%84%E8%A7%A3%E6%9E%90/</link>
      <pubDate>Thu, 03 Sep 2026 00:00:00 +0000</pubDate>
      <guid>https://index.zshipu.com/ai002/post/20260903/32GB%E6%98%BE%E5%AD%98%E5%A6%82%E4%BD%95%E8%B7%91%E9%80%9A56GB%E5%A4%A7%E6%A8%A1%E5%9E%8BShared-Memory%E5%88%B0%E5%BC%82%E6%9E%84%E5%86%85%E5%AD%98%E6%9E%B6%E6%9E%84%E8%A7%A3%E6%9E%90/</guid>
      <description>32GB显存跑56GB大模型，这组数字直接打破了显存必须大于模型参数的常规认知。 文章标题点明核心：Shared Memory与AI异构内存架构是实现路径。前几天一位读者在Juejin发帖，引发对这一技术的讨论。 Shared Memory把系统内存直接映射为显存扩展 Shared Memory机制的核心在于将</description>
    </item>
  </channel>
</rss>
