zero-to-sglangBuild SGLang from Scratch

Start from what inference really is, build a mini-sglang by hand, then read the real SGLang source

Start Learning

简体中文

View on GitHub

Understand inference

KV Cache, prefill vs. decode, compute-bound vs. memory-bound — what these concepts actually mean

Build the engine

Write a mini-sglang from 0 to 1, with every step landing in code

Read the real engine

How RadixAttention, Paged KV Cache and Continuous Batching are implemented inside SGLang

Join open source

Learn profiling and trace analysis, and walk through the full SGLang PR workflow

English edition

Part 0 and Part I are available in English, with further writing and review in progress. Parts II–IV are under development. You can switch to the Chinese edition at any time.