<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>The Minutes of Dylan Patel on multimodal</title>
    <link>https://minutesof.com/dylan-patel/on/multimodal/</link>
    <description>Everything Dylan Patel has said on multimodal: 2 verbatim quotes from October 2025, each with a timestamp and a link to the recording it came from.</description>
    <language>en</language>
    <lastBuildDate>Mon, 31 Aug 2026 13:10:15 +0000</lastBuildDate>
    <atom:link href="https://minutesof.com/dylan-patel/on/multimodal/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Patel says Weka and Vast make high margins on storage for multimodal AI workloads despite drive vendors making</title>
      <link>https://minutesof.com/q/b2443f6f-24bb-433c-a61e-4fea60908146/</link>
      <guid isPermaLink="true">https://minutesof.com/q/b2443f6f-24bb-433c-a61e-4fea60908146/</guid>
      <description>“But then there&#x27;s also, on the storage side, the drive vendors don&#x27;t make any money. But Weka and Vast, I mean, look at their pricing models.” — Open Compute Project</description>
      <pubDate>Thu, 23 Oct 2025 06:01:45 +0000</pubDate>
      <category>multimodal</category>
    </item>
    <item>
      <title>Patel says storage vendors Weka and Vast make high margins on multimodal AI requiring larger storage clusters.</title>
      <link>https://minutesof.com/q/52b91cc2-fa4f-4699-8ca9-06b6a755ef63/</link>
      <guid isPermaLink="true">https://minutesof.com/q/52b91cc2-fa4f-4699-8ca9-06b6a755ef63/</guid>
      <description>“They make crazy margin on storage. And as you get to more multimodal models, image generative image and video robotics, you have much larger storage clusters.” — Open Compute Project</description>
      <pubDate>Thu, 23 Oct 2025 06:01:45 +0000</pubDate>
      <category>multimodal</category>
    </item>
  </channel>
</rss>
