<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>gpu on Paul Piper</title><link>https://www.paul-piper.de/tags/gpu/</link><description>Recent content in gpu on Paul Piper</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><copyright>©Paul Piper</copyright><lastBuildDate>Sat, 06 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://www.paul-piper.de/tags/gpu/index.xml" rel="self" type="application/rss+xml"/><item><title>AIVory Smart Inference - Same AI, Cheaper Bill</title><link>https://www.paul-piper.de/posts/aivory-smart-inference/</link><pubDate>Sat, 06 Jun 2026 00:00:00 +0000</pubDate><guid>https://www.paul-piper.de/posts/aivory-smart-inference/</guid><description>&lt;p>&lt;img
src="https://www.paul-piper.de/images/aivory_black.svg"
alt="AIVory"
width="100"
height="100"
loading="lazy"
decoding="async"
style="float: right; max-width: 250px; margin: 0 0 20px 20px;">&lt;/p>
&lt;p>Okay, so&amp;hellip; we built another product for AIVory: &lt;a href="https://aivory.net/smart-inference" target="_blank" rel="noopener noreferrer">Smart Inference&lt;/a>.&lt;/p>
&lt;p>And the reason is simple: GPU inference is expensive, but most of the time you&amp;rsquo;re overpaying for it. There are $765 billion worth of GPUs out there and roughly half sit idle at any given time. That&amp;rsquo;s why providers offer them at a discount - if you can find their offers and keep track of the actual prices across 60+ providers. That takes time. So we built a service around it.&lt;/p></description></item></channel></rss>