<?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 gpt-5</title>
    <link>https://minutesof.com/dylan-patel/on/gpt-5/</link>
    <description>Everything Dylan Patel has said on gpt-5: 7 verbatim quotes between April 2025 and November 2025, each with a timestamp and a link to the recording it…</description>
    <language>en</language>
    <lastBuildDate>Sun, 30 Aug 2026 16:22:18 +0000</lastBuildDate>
    <atom:link href="https://minutesof.com/dylan-patel/on/gpt-5/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Patel reports Microsoft aims to 10x training capacity every 18-24 months, representing a 10x increase from GPT</title>
      <link>https://minutesof.com/q/fa0ed250-859f-4800-a571-4fb9a5440edb/</link>
      <guid isPermaLink="true">https://minutesof.com/q/fa0ed250-859f-4800-a571-4fb9a5440edb/</guid>
      <description>“We try to 10x the training capacity every eighteen to twenty four months. And so this would be effectively a 10x increase. 10x from what GPD five was trained with.” — Dwarkesh Patel</description>
      <pubDate>Wed, 12 Nov 2025 17:02:10 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel reveals OpenAI&#x27;s major GPT-5 training cluster is in Arizona and runs GPUs at lower power to fit more chi</title>
      <link>https://minutesof.com/q/98e652f5-3e5a-4a75-94a1-443500389683/</link>
      <guid isPermaLink="true">https://minutesof.com/q/98e652f5-3e5a-4a75-94a1-443500389683/</guid>
      <description>“And in some deployments, you&#x27;ll be completely limited on power. And so there&#x27;s many deployments. For example, OpenAI&#x27;s major training cluster in Arizona that did GPT-five and such.” — Open Compute Project</description>
      <pubDate>Thu, 23 Oct 2025 06:01:45 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel reports that OpenAI&#x27;s Orion training run did not improve enough over GPT-4.5 to qualify as GPT-5.</title>
      <link>https://minutesof.com/q/8373ef31-935a-4e8d-90e3-7c6565db225c/</link>
      <guid isPermaLink="true">https://minutesof.com/q/8373ef31-935a-4e8d-90e3-7c6565db225c/</guid>
      <description>“There were hopes that Orion could be used for for GPT five, but its improvement was, like, not enough to be, like, really a GPT five.” — Alex Kantrowitz</description>
      <pubDate>Wed, 23 Apr 2025 16:30:06 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel explains GPT-5 will combine massive pre-training and post-training scale for the first time.</title>
      <link>https://minutesof.com/q/1ac7f4fc-3b6a-4d9e-a854-d3c4009a5164/</link>
      <guid isPermaLink="true">https://minutesof.com/q/1ac7f4fc-3b6a-4d9e-a854-d3c4009a5164/</guid>
      <description>“Along the way, another team at OpenAI made the big breakthrough of reasoning. Right? Strawberry training. And they released o one, and then they released o three.” — Alex Kantrowitz</description>
      <pubDate>Wed, 23 Apr 2025 16:30:06 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel says GPT-5 will combine massive pre-training like GPT-4.5 with massive post-training like o1 and o3.</title>
      <link>https://minutesof.com/q/808fcaf5-9216-420e-b11b-f8a91aeadae8/</link>
      <guid isPermaLink="true">https://minutesof.com/q/808fcaf5-9216-420e-b11b-f8a91aeadae8/</guid>
      <description>“GPT five, as Sam calls it, is is gonna be a model that has huge pre training scale, right, like GPT 4.5, but also huge post training scale like o one and o three and continuing to scale that up.” — Alex Kantrowitz</description>
      <pubDate>Wed, 23 Apr 2025 16:30:06 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel says GPT-5 will be first model to massively scale both pre-training and post-training simultaneously.</title>
      <link>https://minutesof.com/q/00545820-383f-45e6-ac01-d8e99c981b60/</link>
      <guid isPermaLink="true">https://minutesof.com/q/00545820-383f-45e6-ac01-d8e99c981b60/</guid>
      <description>“Right? And this would be the first time we see a model that was was a step up in both at the same time.” — Alex Kantrowitz</description>
      <pubDate>Wed, 23 Apr 2025 16:30:06 +0000</pubDate>
      <category>gpt-5</category>
    </item>
    <item>
      <title>Patel reports GPT-5 is expected in three to six months, possibly sooner based on his sources.</title>
      <link>https://minutesof.com/q/51f3ebb3-3c32-467e-a122-ed7787e9ba81/</link>
      <guid isPermaLink="true">https://minutesof.com/q/51f3ebb3-3c32-467e-a122-ed7787e9ba81/</guid>
      <description>“They say it&#x27;s coming, you know, this year, hopefully, the next three to six months, maybe sooner. I&#x27;ve heard sooner, but, you know, we&#x27;ll we&#x27;ll see.” — Alex Kantrowitz</description>
      <pubDate>Wed, 23 Apr 2025 16:30:06 +0000</pubDate>
      <category>gpt-5</category>
    </item>
  </channel>
</rss>
