---
title: "Papers & systems"
description: "The papers that define modern LLM inference, in reading order: from tiling the attention score matrix to compressing the KV cache away."
canonical_url: "https://fanout.sh/inference-eng/papers"
md_url: "https://fanout.sh/inference-eng/papers.md"
access: "public"
---

# Papers & systems

The papers that define modern LLM inference, in reading order: from tiling the attention score matrix to compressing the KV cache away.

---
This representation contains public Fanout content only. Protected Pro lessons, account data, billing, checkout, and pricing are not included.

Browse the public content map: https://fanout.sh/sitemap.md