<?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>Rag on danilchenko.dev</title><link>https://www.danilchenko.dev/tags/rag/</link><description>Recent content in Rag on danilchenko.dev</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 03 May 2026 08:20:42 +0000</lastBuildDate><atom:link href="https://www.danilchenko.dev/tags/rag/index.xml" rel="self" type="application/rss+xml"/><item><title>MarkItDown vs Docling vs Marker: PDF to Markdown for LLMs</title><link>https://www.danilchenko.dev/posts/markitdown-vs-docling-vs-marker/</link><pubDate>Sun, 03 May 2026 08:20:42 +0000</pubDate><guid>https://www.danilchenko.dev/posts/markitdown-vs-docling-vs-marker/</guid><description>Three open-source PDF-to-Markdown tools for RAG and LLM pipelines, tested on real documents. Speed, table fidelity, install pain, and which one to pick.</description></item><item><title>Recursive Language Models: How RLMs Beat Long Context</title><link>https://www.danilchenko.dev/posts/recursive-language-models/</link><pubDate>Sat, 18 Apr 2026 06:00:00 +0000</pubDate><guid>https://www.danilchenko.dev/posts/recursive-language-models/</guid><description>Recursive language models treat a huge prompt as a Python variable the model can grep and recurse over. MIT&amp;#39;s paper shows it beats GPT-5 on long context.</description></item></channel></rss>