<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Paperless - vo.rs</title><link>https://vo.rs/tags/paperless/</link><description>Latest from the Paperless desk at vo.rs.</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.</copyright><lastBuildDate>Sun, 22 Sep 2024 09:19:00 +0000</lastBuildDate><atom:link href="https://vo.rs/tags/paperless/" rel="self" type="application/rss+xml"/><item><title>Paperless + a Local LLM: Auto-Tagging Scans Offline</title><link>https://vo.rs/story/paperless-a-local-llm-auto-tagging-scans-offline/</link><description>&lt;p&gt;Paperless-ngx solved the boring half of my paperwork years ago. I feed it scans, it runs OCR, and every invoice, letter and warranty becomes full-text searchable in a tidy web interface. What it never solved was the tedious half: putting each document in the right place. A fresh scan lands with no tags, no correspondent, and a filename like &lt;code&gt;scan_0042.pdf&lt;/code&gt;, and unless I stop and classify it by hand, my archive slowly rots into a searchable-but-unsorted heap. For a while I did the classifying. Then I got a local language model reading the OCR text and doing it for me, offline, and I stopped touching the pile entirely.&lt;/p&gt;</description><pubDate>Sun, 22 Sep 2024 09:19:00 +0000</pubDate></item></channel></rss>