<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Model-Routing on Tokenise</title><link>https://tokenise.rosvetic.com/tags/model-routing/</link><description>Recent content in Model-Routing on Tokenise</description><generator>Hugo</generator><language>en-gb</language><lastBuildDate>Thu, 05 Nov 2026 11:07:00 +0000</lastBuildDate><atom:link href="https://tokenise.rosvetic.com/tags/model-routing/index.xml" rel="self" type="application/rss+xml"/><item><title>Copilot Fusion picks the model for you. Measure what it costs per accepted change</title><link>https://tokenise.rosvetic.com/posts/copilot-fusion-ga-measure-on-your-own-tasks/</link><pubDate>Thu, 05 Nov 2026 11:07:00 +0000</pubDate><guid>https://tokenise.rosvetic.com/posts/copilot-fusion-ga-measure-on-your-own-tasks/</guid><description>&lt;p&gt;GitHub made Copilot Fusion generally available on 28 October, and the pitch is simple: stop choosing a model per task, pick Fusion and an effort level, and let Copilot route. &lt;a href="https://github.blog/changelog/2026-10-28-route-and-orchestrate-models-with-github-copilot-fusion" target="_blank" rel="noopener noreferrer"&gt;The changelog&lt;/a&gt;&#10; puts a number on it. At the Max level, Fusion scored 60 on the Artificial Analysis Coding Agent Index, within two points of Claude Code running Opus 5.5 at max effort, for about half the average API cost per task.&lt;/p&gt;</description></item></channel></rss>