<?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>VLLM on ITNote 技术笔记</title>
    <link>https://www.itnote.org/tags/vllm/</link>
    <description>Recent content in VLLM on ITNote 技术笔记</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Mon, 27 Jul 2026 09:20:00 +0800</lastBuildDate>
    <atom:link href="https://www.itnote.org/tags/vllm/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>基于 vLLM 搭建 OpenAI 兼容 API 服务并接入现有应用</title>
      <link>https://www.itnote.org/posts/vllm-openai-api-service/</link>
      <pubDate>Mon, 27 Jul 2026 09:20:00 +0800</pubDate>
      <guid>https://www.itnote.org/posts/vllm-openai-api-service/</guid>
      <description>从零搭建一个可被 LangChain、OpenAI SDK 等直接调用的本地推理服务，并配置进程守护与反向代理。</description>
    </item>
    <item>
      <title>vLLM 显存优化与批处理参数调优实战</title>
      <link>https://www.itnote.org/posts/vllm-memory-tuning/</link>
      <pubDate>Mon, 27 Jul 2026 09:10:00 +0800</pubDate>
      <guid>https://www.itnote.org/posts/vllm-memory-tuning/</guid>
      <description>详解 vLLM 中影响显存占用与吞吐量的关键参数，帮助在有限硬件资源下获得最佳推理性能。</description>
    </item>
    <item>
      <title>vLLM 入门：轻量级大模型推理引擎原理与部署</title>
      <link>https://www.itnote.org/posts/vllm-intro/</link>
      <pubDate>Mon, 27 Jul 2026 09:00:00 +0800</pubDate>
      <guid>https://www.itnote.org/posts/vllm-intro/</guid>
      <description>介绍 vLLM 的核心设计理念 PagedAttention，以及在单机环境下快速部署一个轻量级推理服务的完整流程。</description>
    </item>
  </channel>
</rss>
