The most-installed JS frameworks aren't the most-starred — I checked 14
GitHub stars are the number everyone quotes, but they measure admiration, not usage. So I pulled two public signals for the 14 most-tracked JS framework tools — GitHub stars (mindshare) and npm weekly downloads (actual u
GitHub stars are the number everyone quotes, but they measure admiration, not
usage. So I pulled two public signals for the 14 most-tracked JS framework tools
— GitHub stars (mindshare) and npm weekly downloads (actual usage) — and
lined them up. The gap is bigger than I expected, and it flips the leaderboard.
Every number here is a live public signal captured on 2026-06-15. No private
data, no scraping behind logins.
The leaderboard (ranked by GitHub stars)
| Tool | GitHub stars | npm / week |
|---|---|---|
| Next.js | 140.0k | 37.9M |
| Angular | 100.3k | 6.1M |
| Svelte | 87.3k | 4.8M |
| NestJS | 75.9k | 11.4M |
| Nuxt | 60.4k | 1.7M |
| Astro | 60.2k | 3.2M |
| Vue | 53.8k | 13.7M |
| TanStack Query | 49.8k | 58.5M |
| Fastify | 36.5k | 8.4M |
| SolidJS | 35.6k | 2.6M |
| Remix | 33.1k | 791.2k |
| Hono | 31.0k | 44.8M |
| Qwik | 22.0k | 29.7k |
| SvelteKit | 20.6k | 2.1M |
The part that surprised me
Two tools out-install Next.js while carrying a fraction of its stars:
- TanStack Query — 49.8k stars, but 58.5M installs/week
- Hono — 31.0k stars, but 44.8M installs/week
- Next.js — 140.0k stars, 37.9M installs/week
Both are drop-in libraries you add to a project and ship — you rarely stop to
star them. The big frameworks, meanwhile, rack up stars from people who admire
them but never deploy with them.
You can see it clearly in downloads-per-star (a rough "how much do people
actually use this per unit of hype"):
- Hono — 1,445x
- TanStack Query — 1,175x
- Next.js — 271x
- Vue — 255x
- Svelte — 55x
- Remix — 24x
A ~60x spread. Installs also concentrate hard at the top: the three biggest
(TanStack Query, Hono, Next.js) are about 72% of all weekly downloads in this
set, while Qwik barely registers at ~30k/week.
The takeaway
Stars are a lagging, vanity-adjacent indicator. If you're picking a tool — or
sizing up a competitor — weekly installs, release cadence and who's hiring tell
you far more about real momentum than a star count that took five years to
accumulate.
I put this together while building Sigmon, a tool that
tracks dev-tool competitors across public signals (GitHub, npm, releases,
hiring, funding) and sends a weekly brief of what changed. The full interactive
leaderboard with head-to-head comparisons is
here.
Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.