<?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>Rust on Ragib CS</title>
    <link>https://ragibcs.gitlab.io/tags/rust/</link>
    <description>Recent content in Rust on Ragib CS</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Fri, 03 Apr 2026 10:00:00 +0600</lastBuildDate>
    <atom:link href="https://ragibcs.gitlab.io/tags/rust/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Building linfer: A Rust-Based LLM Inference Engine</title>
      <link>https://ragibcs.gitlab.io/posts/building-linfer-rust-llm-inference/</link>
      <pubDate>Fri, 03 Apr 2026 10:00:00 +0600</pubDate>
      <guid>https://ragibcs.gitlab.io/posts/building-linfer-rust-llm-inference/</guid>
      <description>&lt;h2 id=&#34;why-build-another-inference-engine&#34;&gt;Why Build Another Inference Engine?&lt;/h2&gt;
&lt;p&gt;When I started working with large language models locally, I quickly ran into the usual suspects: slow inference times, memory bloat, and dependency hell. Most existing solutions are either too heavyweight (PyTorch with CUDA) or too opinionated about model formats.&lt;/p&gt;
&lt;p&gt;So I built &lt;strong&gt;linfer&lt;/strong&gt; — a Rust-based local LLM inference engine that&amp;rsquo;s 3x faster than comparable solutions for CPU inference.&lt;/p&gt;
&lt;h2 id=&#34;the-core-problem&#34;&gt;The Core Problem&lt;/h2&gt;
&lt;p&gt;Running LLMs locally on CPU is painful:&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
