<?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 Cognaptus</title>
    <link>https://cognaptus.com/tags/vllm/</link>
    <description>Recent content in VLLM on Cognaptus</description>
    <generator>Hugo -- 0.145.0</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 27 May 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cognaptus.com/tags/vllm/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The Experts Are Sparse Inside: Why MoE Cost Cuts Stop at 1.2x</title>
      <link>https://cognaptus.com/blog/2026-05-27-the-experts-are-sparse-inside-why-moe-cost-cuts-stop-at-12x/</link>
      <pubDate>Wed, 27 May 2026 00:00:00 +0000</pubDate>
      <guid>https://cognaptus.com/blog/2026-05-27-the-experts-are-sparse-inside-why-moe-cost-cuts-stop-at-12x/</guid>
      <description>A mechanism-first reading of intra-expert activation sparsity in MoE models, and why large theoretical sparsity becomes modest but useful inference savings in production.</description>
    </item>
  </channel>
</rss>
