<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Models on Aguilera Engineering</title><link>https://aguilera.ee/blog/tags/models/</link><description>Recent content in Models on Aguilera Engineering</description><generator>Hugo -- gohugo.io</generator><language>en-US</language><managingEditor>eduardo@aguilera.ee (Eduardo Aguilera)</managingEditor><webMaster>eduardo@aguilera.ee (Eduardo Aguilera)</webMaster><copyright>Eduardo Aguilera</copyright><lastBuildDate>Wed, 07 Jan 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://aguilera.ee/blog/tags/models/index.xml" rel="self" type="application/rss+xml"/><item><title>The cost of the best models</title><link>https://aguilera.ee/blog/cost-of-the-best-models/</link><pubDate>Wed, 07 Jan 2026 00:00:00 +0000</pubDate><author>eduardo@aguilera.ee (Eduardo Aguilera)</author><guid>https://aguilera.ee/blog/cost-of-the-best-models/</guid><description>&lt;p&gt;The best models, per seat:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;ChatGPT Business: 30 dollars&lt;/li&gt;
&lt;li&gt;Claude Pro: 20 dollars&lt;/li&gt;
&lt;li&gt;Gemini Enterprise: 21 dollars&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;Unless you&amp;rsquo;re on a Business plan, OpenAI and Gemini train on your data. More on
that &lt;a href="https://aguilera.ee/blog/openai-trains-on-your-data/"&gt;here&lt;/a&gt;.&lt;/p&gt;
&lt;p&gt;That&amp;rsquo;s 71 dollars per person on starter plans, and it climbs the moment you
exhaust the quotas. Expensive, and hard to control securely at scale.&lt;/p&gt;
&lt;p&gt;One subscription, all three, or something else entirely is a real decision.&lt;/p&gt;</description><content:encoded><![CDATA[<p>The best models, per seat:</p>
<ul>
<li>ChatGPT Business: 30 dollars</li>
<li>Claude Pro: 20 dollars</li>
<li>Gemini Enterprise: 21 dollars</li>
</ul>
<p>Unless you&rsquo;re on a Business plan, OpenAI and Gemini train on your data. More on
that <a href="/blog/openai-trains-on-your-data/">here</a>.</p>
<p>That&rsquo;s 71 dollars per person on starter plans, and it climbs the moment you
exhaust the quotas. Expensive, and hard to control securely at scale.</p>
<p>One subscription, all three, or something else entirely is a real decision.</p>
]]></content:encoded></item><item><title>Is OpenAI training on your data?</title><link>https://aguilera.ee/blog/openai-trains-on-your-data/</link><pubDate>Tue, 06 Jan 2026 00:00:00 +0000</pubDate><author>eduardo@aguilera.ee (Eduardo Aguilera)</author><guid>https://aguilera.ee/blog/openai-trains-on-your-data/</guid><description>&lt;p&gt;Is OpenAI training its models on your data?&lt;/p&gt;
&lt;p&gt;Short answer: yes. On the Free, Go, Plus, and Pro plans, your data is used for
training. On Business and Enterprise, it isn&amp;rsquo;t.&lt;/p&gt;
&lt;p&gt;Worth knowing before you hand ChatGPT subscriptions to your team.&lt;/p&gt;</description><content:encoded><![CDATA[<p>Is OpenAI training its models on your data?</p>
<p>Short answer: yes. On the Free, Go, Plus, and Pro plans, your data is used for
training. On Business and Enterprise, it isn&rsquo;t.</p>
<p>Worth knowing before you hand ChatGPT subscriptions to your team.</p>
]]></content:encoded></item><item><title>Stop using the wrong AI for the job</title><link>https://aguilera.ee/blog/wrong-ai-for-the-job/</link><pubDate>Thu, 25 Dec 2025 00:00:00 +0000</pubDate><author>eduardo@aguilera.ee (Eduardo Aguilera)</author><guid>https://aguilera.ee/blog/wrong-ai-for-the-job/</guid><description>&lt;p&gt;Not every model trains on the same data. Each one leads its own category.
Berkeley&amp;rsquo;s LMArena lets you compare models side by side and casts votes on a
public leaderboard.&lt;/p&gt;
&lt;p&gt;The leaders in December 2025:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Text: Gemini 3 Pro&lt;/li&gt;
&lt;li&gt;Coding: Claude Opus 4.5&lt;/li&gt;
&lt;li&gt;Text-to-image: GPT Image 1.5&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;I asked Claude and Gemini for a title for this post. Gemini was thorough: three
options across three angles, with a recommendation. Claude gave me half as many
and no reasoning.&lt;/p&gt;</description><content:encoded><![CDATA[<p>Not every model trains on the same data. Each one leads its own category.
Berkeley&rsquo;s LMArena lets you compare models side by side and casts votes on a
public leaderboard.</p>
<p>The leaders in December 2025:</p>
<ul>
<li>Text: Gemini 3 Pro</li>
<li>Coding: Claude Opus 4.5</li>
<li>Text-to-image: GPT Image 1.5</li>
</ul>
<p>I asked Claude and Gemini for a title for this post. Gemini was thorough: three
options across three angles, with a recommendation. Claude gave me half as many
and no reasoning.</p>
<p>Claude&rsquo;s suggestion was &ldquo;Are You Using the Wrong AI?&rdquo;. Gemini&rsquo;s is the title that
kept you reading.</p>
]]></content:encoded></item></channel></rss>