<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>The Koleslaw Blog</title>
    <link>https://koleslaw.ai/blog</link>
    <description>Notes from running a fine-tuned 30B model in production: quantization, spot GPUs, failure design, and what it all costs.</description>
    <language>en</language>
    <atom:link href="https://koleslaw.ai/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>My &quot;4-bit&quot; quant was 6.2 bits per weight</title>
      <link>https://koleslaw.ai/blog/the-quantization-that-didnt-fit</link>
      <guid isPermaLink="true">https://koleslaw.ai/blog/the-quantization-that-didnt-fit</guid>
      <pubDate>Wed, 29 Jul 2026 00:00:00 GMT</pubDate>
      <description>Nemotron&#39;s hidden dimension is 2688, which is not divisible by 256. That one fact quietly inflated a 4-bit quant by a third and cost a weekend.</description>
    </item>
  </channel>
</rss>
