<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>DeepSeek on The Coders Blog</title><link>https://thecodersblog.com/tag/deepseek/</link><description>Recent content in DeepSeek on The Coders Blog</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 07 May 2026 21:08:12 +0000</lastBuildDate><atom:link href="https://thecodersblog.com/tag/deepseek/index.xml" rel="self" type="application/rss+xml"/><item><title>DeepSeek 4 Flash: Local LLM Inference on Metal</title><link>https://thecodersblog.com/deepseek-4-flash-local-inference-engine-2026/</link><pubDate>Thu, 07 May 2026 21:08:12 +0000</pubDate><guid>https://thecodersblog.com/deepseek-4-flash-local-inference-engine-2026/</guid><description>&lt;p&gt;Forget the cloud. The future of powerful AI is landing squarely on your desk, and with DeepSeek 4 Flash, it&amp;rsquo;s running blazing fast on your Mac. Salvatore Sanfilippo, the architect behind Redis, has delivered &lt;code&gt;ds4.c&lt;/code&gt;, a remarkably specialized inference engine designed exclusively for the DeepSeek V4 Flash model, and crucially, for Apple Silicon&amp;rsquo;s Metal GPU. This isn&amp;rsquo;t just another &lt;code&gt;llama.cpp&lt;/code&gt; clone; it&amp;rsquo;s a laser-focused piece of engineering democratizing on-device AI.&lt;/p&gt;</description></item></channel></rss>