thinQmaster Virtual testing partner Adaptive practice exams that find your weak spots and drill them. Every answer explained. From $69
Enroll

thinQmaster

Find your exam

Five questions free, before you decide.

Step 1 · choose a vendor

In one of our camps? Your practice exams come with the tuition — they are on your study page already.

AI & Machine Learning · Advanced

PagedAttention

A KV-cache memory technique, introduced by vLLM, that stores cache in fixed-size blocks like virtual memory pages.

Definition

Paging avoids fragmentation and over-allocation, letting servers fit many more concurrent sequences and share prefixes. It is a core reason for vLLM's high throughput.

Exam tip

PagedAttention raises concurrency by reducing KV-cache memory waste, not by changing model accuracy.

Related terms