<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Inference on Michael Barrow</title><link>https://blog.barrow.me/tags/inference/</link><description>Recent content in Inference on Michael Barrow</description><generator>Hugo</generator><language>en-us</language><copyright>Michael Barrow</copyright><lastBuildDate>Sat, 03 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.barrow.me/tags/inference/index.xml" rel="self" type="application/rss+xml"/><item><title>These are not the models you're looking for</title><link>https://blog.barrow.me/posts/these-are-not-the-models-youre-looking-for/</link><pubDate>Sat, 03 Oct 2026 00:00:00 +0000</pubDate><guid>https://blog.barrow.me/posts/these-are-not-the-models-youre-looking-for/</guid><description>I read the LLM benchmark charts, but they don&amp;rsquo;t tell me what I need to know. What does: data terms, billing shape, reliability, and starting cheap before escalating.</description></item></channel></rss>