<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>张量并行 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E5%BC%A0%E9%87%8F%E5%B9%B6%E8%A1%8C/</link><description>Recent content in 张量并行 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Mon, 13 Jul 2026 09:00:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E5%BC%A0%E9%87%8F%E5%B9%B6%E8%A1%8C/index.xml" rel="self" type="application/rss+xml"/><item><title>大模型训练的分布式优化策略：从数据并行到3D并行</title><link>https://guijiagi.com/posts/article-97/</link><pubDate>Mon, 13 Jul 2026 09:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/article-97/</guid><description>系统解析大模型分布式训练的并行策略，涵盖数据并行、张量并行、流水线并行及混合策略</description></item><item><title>张量并行详解：Megatron-LM</title><link>https://guijiagi.com/posts/tensor-parallelism-deep/</link><pubDate>Thu, 02 Jul 2026 11:05:00 +0800</pubDate><guid>https://guijiagi.com/posts/tensor-parallelism-deep/</guid><description>从原理到实现，深入解析Megatron-LM张量并行方案的设计与工程细节</description></item></channel></rss>