<?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>GPU Serving on Cognaptus</title>
    <link>https://cognaptus.com/tags/gpu-serving/</link>
    <description>Recent content in GPU Serving on Cognaptus</description>
    <generator>Hugo -- 0.145.0</generator>
    <language>en-us</language>
    <lastBuildDate>Sun, 09 Aug 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cognaptus.com/tags/gpu-serving/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Busy Compute, Idle Memory: ServerlessT2I Rebuilds Image Serving Around Model DAGs</title>
      <link>https://cognaptus.com/blog/2026-08-09-busy-compute-idle-memory-serverlesst2i-rebuilds-image-serving-around-model-dags/</link>
      <pubDate>Sun, 09 Aug 2026 00:00:00 +0000</pubDate>
      <guid>https://cognaptus.com/blog/2026-08-09-busy-compute-idle-memory-serverlesst2i-rebuilds-image-serving-around-model-dags/</guid>
      <description>ServerlessT2I shows how model-granular workflow serving can turn unused GPU memory into higher image-generation capacity, tighter SLOs, and more defensible tenant accounting.</description>
    </item>
  </channel>
</rss>
