Subject areaData and infrastructure
llms.txt is a plain-text file placed at the root of a website (at /llms.txt) to introduce the site to large language models, or LLMs (the systems behind AI assistants).
It is a proposal published in 2024 by Jeremy Howard, and it remains a voluntary convention with no body governing it.
The file is written in Markdown (a way of formatting text with a few symbols, readable by a person too): a heading with the site name, a short summary and then lists of the pages that matter, each with one line of explanation. HTML pages are often published alongside Markdown twins, free of menus, ads and code, so the model reads the content and spends fewer resources cleaning it up.
It complements the other files at the root: robots.txt tells crawlers what they may visit, the sitemap lists every address, and llms.txt points to the pages that explain the site best and the order in which to read them.
An example
The case below is constructed, because the term does not yet appear in the articles. A financial advisory firm puts its name, two lines on what it does and the list of its service pages in /llms.txt, each linked to its Markdown version: an assistant answering a question about its sector finds clean text to quote right away.
Enrico Giubertoni. Licensed under CC BY-NC 4.0: you may quote, translate and adapt this entry with the author’s name and a link, stating any changes. For commercial use, write to info@enricogiubertoni.com
Spotted an inaccuracy or something that needs updating? Write to info@enricogiubertoni.com and help make this glossary a better place.
What are your business challenges?
Tell us about your priorities and the objectives you want to reach.
You will receive a free, targeted answer within one working day.