<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Quantization on Dustin Morgan</title><link>https://dustinmorgan.org/tags/quantization/</link><description>Recent content in Quantization on Dustin Morgan</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 26 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://dustinmorgan.org/tags/quantization/index.xml" rel="self" type="application/rss+xml"/><item><title>What I Got Wrong About SQ4</title><link>https://dustinmorgan.org/blog/what-i-got-wrong-about-sq4/</link><pubDate>Wed, 26 Aug 2026 00:00:00 +0000</pubDate><guid>https://dustinmorgan.org/blog/what-i-got-wrong-about-sq4/</guid><description>I published a 4-bit quantization format and called its quality indistinguishable from FP16. Once I measured it properly, standard 4-bit formats beat it. I&amp;rsquo;ve retired it.</description></item><item><title>SQ4 Is Live: Whitepaper + llama.cpp PR</title><link>https://dustinmorgan.org/blog/sq4-is-live-llama-cpp-pr-and-whitepaper/</link><pubDate>Wed, 13 May 2026 00:00:00 +0000</pubDate><guid>https://dustinmorgan.org/blog/sq4-is-live-llama-cpp-pr-and-whitepaper/</guid><description>We published the SQ4 whitepaper and opened a PR to bring percentile-calibrated 4-bit quantization to llama.cpp.</description></item></channel></rss>