<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title>Blog on Just One Notebook</title>
		<link>https://justonenotebook.com/blog/</link>
		<description>Recent content in Blog on Just One Notebook</description>
		<generator>Hugo</generator>
		<language>en-US</language>
		
		
		
		
			<atom:link href="https://justonenotebook.com/blog/index.xml" rel="self" type="application/rss+xml" />
			<item>
				<title>Why is ARC-AGI interesting?</title>
				<link>https://justonenotebook.com/blog/arc-agi/</link>
				<pubDate>Fri, 07 Aug 2026 00:00:00 +0000</pubDate>
				<guid>https://justonenotebook.com/blog/arc-agi/</guid>
				<description>&lt;p&gt;&lt;em&gt;Disclaimer: I am competing in both the ARC-AGI-2 and ARC-AGI-3 competitions on Kaggle, and I intend to do as well as I can.&#xA;I am not affiliated with &lt;a href=&#34;https://arcprize.org/&#34;&gt;ARC Prize&lt;/a&gt; in any other way.&lt;/em&gt;&lt;/p&gt;&#xA;&lt;h2 id=&#34;its-saturated-already&#34;&gt;It&amp;rsquo;s Saturated Already?&lt;a class=&#34;heading-anchor&#34; href=&#34;#its-saturated-already&#34; aria-hidden=&#34;true&#34; tabindex=&#34;-1&#34;&gt;#&lt;/a&gt;&lt;/h2&gt;&#xA;&lt;p&gt;ARC-AGI-1/2 have been saturated by frontier LLMs (see the caveat about benchmaxxing below).&lt;/p&gt;&#xA;&lt;p&gt;The next interesting question is cost per task.&#xA;&lt;a href=&#34;https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731&#34;&gt;DeepSeek-V4-Flash-0731&lt;/a&gt;, for example, is far ahead of the pack in its price range.&lt;/p&gt;</description>
			</item>
	</channel>
</rss>
