<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Llamacpp on Rost Glukhov | Personal site and technical blog</title>
		<link>https://www.glukhov.org/tags/llamacpp/</link>
		<description>Recent content in Llamacpp on Rost Glukhov | Personal site and technical blog</description>
		<generator>Hugo</generator>
		<language>en-us</language>
		
		
		
		
			<lastBuildDate>Tue, 08 Sep 2026 00:00:00 +0000</lastBuildDate>
		
			<atom:link href="https://www.glukhov.org/tags/llamacpp/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>KV Cache on 16 GB GPUs: Making Long Context Actually Fit</title>
				<link>https://www.glukhov.org/llm-performance/optimization/kv-cache-16gb-long-context/</link>
				<pubDate>Tue, 08 Sep 2026 00:00:00 +0000</pubDate>
				<guid>https://www.glukhov.org/llm-performance/optimization/kv-cache-16gb-long-context/</guid>
				<description>&lt;p&gt;A model can advertise a 128K context window and still fail at 40K tokens on a 16 GB GPU. The architecture ceiling never promised that weights, KV cache, compute buffers, and the desktop compositor would fit on your card at the same time.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
