<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Benchmarks on Fulloch</title><link>https://fulloch.com/tags/benchmarks/</link><description>Recent content in Benchmarks on Fulloch</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 Liam Pettigrew</copyright><lastBuildDate>Thu, 18 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://fulloch.com/tags/benchmarks/index.xml" rel="self" type="application/rss+xml"/><item><title>Watching a Voice Assistant Get Dumber: Qwen3.5 from 9B Down to 0.8B</title><link>https://fulloch.com/posts/watching-a-voice-assistant-get-dumber/</link><pubDate>Thu, 18 Jun 2026 00:00:00 +0000</pubDate><guid>https://fulloch.com/posts/watching-a-voice-assistant-get-dumber/</guid><description>Same conversation, same prompt, same tools, four model sizes. Watching Fulloch degrade from a capable agent into something that confuses the World Cup with Formula 1.</description></item></channel></rss>