<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Why the first load is slow and every one after is fast]]></title><description><![CDATA[<p dir="auto">A question that comes up a lot: <em>"the first time I load a model it takes a while, then it's instant. Why?"</em></p>
<p dir="auto">The engine compiles small GPU programs (shaders) for your specific hardware the first time it runs. That compile used to happen on <strong>every</strong> load, which meant waiting a minute or more each time.</p>
<p dir="auto">Those compiled shaders are now written to disk and reused. So:</p>
<ul>
<li><strong>First ever load on this machine:</strong> a few seconds of compiling.</li>
<li><strong>Every load after:</strong> near-instant, including different models and after a reboot.</li>
</ul>
<p dir="auto">If your first load is still slow every single time, that's worth reporting — it usually means the cache directory isn't writable.</p>
<p dir="auto">How long does a load take on your setup, cold vs. warm?</p>
]]></description><link>https://community.lizard-llm.qendryx.com/topic/4/why-the-first-load-is-slow-and-every-one-after-is-fast</link><generator>RSS for Node</generator><lastBuildDate>Mon, 24 Aug 2026 03:19:02 GMT</lastBuildDate><atom:link href="https://community.lizard-llm.qendryx.com/topic/4.rss" rel="self" type="application/rss+xml"/><pubDate>Mon, 13 Jul 2026 10:16:00 GMT</pubDate><ttl>60</ttl></channel></rss>