<?xml version="1.0" encoding="UTF-8" ?><rss version="2.0"><channel><title>Thinking Tiny Thoughts Devlog - itch.io</title><link>https://southscribblecompany.itch.io/thinking-tiny-thoughts</link><description>Fully offline, zero-data-leak local AI powered by Qwen2.5, raylib, and on-device CPU inference.</description><item><guid>https://southscribblecompany.itch.io/thinking-tiny-thoughts/devlog/1680084/pure-c-transformer-inference</guid><title>Pure C Transformer Inference</title><link>https://southscribblecompany.itch.io/thinking-tiny-thoughts/devlog/1680084/pure-c-transformer-inference</link><pubDate>Sun, 27 Sep 2026 23:27:46 GMT</pubDate><category>devlog</category><description><![CDATA[<img class="post_image" src="https://img.itch.zone/aW1nLzMwMzI0MjQxLmpwZw==/400x224%23c/veyvk4.jpg"/><p>Technical Deep Dive: Transformer Inference When engineering Thinking Tiny Thoughts , our primary objective was to demonstrate that real-time, high-quality conversational generative AI can run complete...</p>]]></description></item></channel></rss>