// ⚡ Inference Acceleration & Serving
PagedAttention (vLLM)
Manages KV cache in pages like OS virtual memory.
PagedAttention（vLLM） 2023 Intermediate core
IN PLAIN WORDS
Split each request's KV cache into fixed-size blocks allocated on demand and shareable, instead of reserving contiguous memory. Fragmentation nearly vanishes and throughput rises several-fold.
Used in: vLLM and most serving engines
01Original paper
Efficient Memory Management for Large Language Model Serving with PagedAttention
Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, et al. · 2023
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Related:Continuous (Iteration-Level) Batching
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
04Where the course teaches it
On learning paths:AI Application Engineer
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
paged-attention)