<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>MATRIX · Liquid d1</title>
  <link>https://derekwang2002.github.io/ai-intelligence/en/projects/liquid-d1/</link>
  <description>Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations.</description>
  <language>en</language>
  <lastBuildDate>Mon, 02 Nov 2026 16:56:00 GMT</lastBuildDate>
  <atom:link href="https://derekwang2002.github.io/ai-intelligence/en/feeds/projects/liquid-d1.xml" rel="self" type="application/rss+xml"/>
  <item>
    <title>Liquid AI launches the hosted d1-ultra decision API</title>
    <link>https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261007-02/</link>
    <guid isPermaLink="false">ev-20261007-02-d1-ultra</guid>
    <pubDate>Fri, 16 Oct 2026 15:56:07 GMT</pubDate>
    <category>Update</category>
    <category>TRIAL</category>
    <description>&lt;p&gt;Liquid AI launches the hosted d1-ultra decision API&lt;/p&gt;&lt;p&gt;Original story: Liquid AI releases multimodal d1 decision models for edge devices&lt;/p&gt;&lt;p&gt;Recommendation: TRIAL&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://www.liquid.ai/blog/d1-ultra&quot;&gt;Liquid d1-ultra&lt;/a&gt;&lt;/p&gt;</description>
  </item>
  <item>
    <title>Liquid AI releases multimodal d1 decision models for edge devices</title>
    <link>https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261007-02/</link>
    <guid isPermaLink="false">ev-20261007-02-published</guid>
    <pubDate>Wed, 07 Oct 2026 22:04:39 GMT</pubDate>
    <category>New</category>
    <category>TRIAL</category>
    <description>&lt;p&gt;Liquid AI now offers both downloadable d1 decision weights and the hosted multimodal d1-ultra API. d1-ultra returns fixed-choice probabilities for text and image decisions at $0.042 per million input tokens with no output-token charge. Vendor evaluations across 112 tasks report gains over Jev 1.14; independent results are not yet available.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Why read: &lt;/strong&gt;Liquid d1 moves routing and review from token generation to one structured decision, warranting latency and calibration tests.&lt;/p&gt;&lt;p&gt;Recommendation: TRIAL&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://huggingface.co/blog/LiquidAI/open-d1&quot;&gt;huggingface.co · open d1&lt;/a&gt; · &lt;a href=&quot;https://huggingface.co/LiquidAI/d1-3B&quot;&gt;huggingface.co · d1 3B&lt;/a&gt;&lt;/p&gt;</description>
  </item>
</channel>
</rss>
