<?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>显存优化 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E6%98%BE%E5%AD%98%E4%BC%98%E5%8C%96/</link><description>Recent content in 显存优化 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>Thu, 02 Jul 2026 11:04:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E6%98%BE%E5%AD%98%E4%BC%98%E5%8C%96/index.xml" rel="self" type="application/rss+xml"/><item><title>梯度检查点原理与实现</title><link>https://guijiagi.com/posts/gradient-checkpointing-explained/</link><pubDate>Thu, 02 Jul 2026 11:04:00 +0800</pubDate><guid>https://guijiagi.com/posts/gradient-checkpointing-explained/</guid><description>解析梯度检查点如何以时间换空间，使超大模型训练成为可能</description></item><item><title>KV Cache优化全攻略：从PagedAttention到MLA</title><link>https://guijiagi.com/posts/kv-cache-optimization-guide/</link><pubDate>Tue, 30 Jun 2026 11:50:00 +0800</pubDate><guid>https://guijiagi.com/posts/kv-cache-optimization-guide/</guid><description>全面解析KV Cache优化技术，从PagedAttention到MLA的深度对比</description></item><item><title>QLoRA量化微调实战：显存减半效果不减</title><link>https://guijiagi.com/posts/qlora-quantized-finetuning-guide/</link><pubDate>Tue, 30 Jun 2026 09:40:00 +0800</pubDate><guid>https://guijiagi.com/posts/qlora-quantized-finetuning-guide/</guid><description>QLoRA让7B模型微调只需8GB显存，70B模型只需24GB，本文详解QLoRA原理与实战配置</description></item></channel></rss>