<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Review on My New Hugo Project</title><link>https://example.org/en/tags/review/</link><description>Recent content in Review on My New Hugo Project</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 Blog Author</copyright><lastBuildDate>Sun, 28 Dec 2025 10:00:00 +0800</lastBuildDate><atom:link href="https://example.org/en/tags/review/index.xml" rel="self" type="application/rss+xml"/><item><title>Tech Year in Review 2025</title><link>https://example.org/en/posts/tech-review-2025/</link><pubDate>Sun, 28 Dec 2025 10:00:00 +0800</pubDate><guid>https://example.org/en/posts/tech-review-2025/</guid><description>&lt;h2 class="relative group"&gt;LLMs move from demo to production&#10; &lt;div id="llms-move-from-demo-to-production" class="anchor"&gt;&lt;/div&gt;&#10; &#10; &lt;span&#10; class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none"&gt;&#10; &lt;a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#llms-move-from-demo-to-production" aria-label="Anchor"&gt;#&lt;/a&gt;&#10; &lt;/span&gt;&#10; &#10;&lt;/h2&gt;&#10;&lt;p&gt;In 2025, large language models left the &amp;ldquo;toy&amp;rdquo; stage and entered real business scenarios: code completion, customer support, and document processing. &lt;strong&gt;RAG and agents&lt;/strong&gt; became the mainstream pattern for enterprise adoption, while inference costs kept falling.&lt;/p&gt;</description></item></channel></rss>