An SSG benchmark across five React frameworks, from one thousand

pages to half a million.

You're building a marketplace. Or a documentation site. A wiki,

a generated archive, any of a dozen things that ship a static

catalogue at scale. Your CMS has a hundred thousand entries.

You've picked your SSG. You run the build.

Five minutes. Ten. Twenty. Maybe an hour. Maybe a stack trace.

You don't know in advance — and the public benchmarks won't tell

you. Most stop at a thousand pages, where most real catalogues

start. The gap between what gets measured and what gets shipped

is where the unpleasant surprises live, and the engineer who has

to ship into that gap usually finds out which side of it their

tool was designed for at deploy time.

So I built a benchmark for the gap.

The benchmark

Five frameworks in a pnpm workspace, each rendering one dynamic

route /posts/[id] from a shared deterministic data source. Same

content, same shape, idiomatic config per tool.…