<?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>Pytorch on Aillex — Do-It-Yourself AI</title>
    <link>https://askaillex.com/tags/pytorch/</link>
    <description>Recent content in Pytorch on Aillex — Do-It-Yourself AI</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Sat, 14 Nov 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://askaillex.com/tags/pytorch/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Installed PyTorch for CUDA 12.8? ComfyUI Is Skipping Its Fast Kernels.</title>
      <link>https://askaillex.com/articles/radar-cu128-comfyui-fast-kernels-off/</link>
      <pubDate>Sat, 14 Nov 2026 00:00:00 +0000</pubDate>
      <guid>https://askaillex.com/articles/radar-cu128-comfyui-fast-kernels-off/</guid>
      <description>A 5090 owner got a 20-second MiniMax H3 clip that ran out of memory to fit, just by reinstalling PyTorch for a newer CUDA. That is one person&amp;rsquo;s result. The reason behind it is in ComfyUI&amp;rsquo;s own code, it applies to every recent NVIDIA card, and you can check your install in under a minute.</description>
    </item>
  </channel>
</rss>
