Weight is the headline number, but the count of separate things a browser must fetch is what people feel on a phone. Paste the page source below — or drop the file in — and this reports both, plus what blocks the first render.
.html file here| Things the browser must fetch | — |
| Of those, render-blocking | — |
| Third-party hosts | — |
Images without loading="lazy" | — |
| Iframes | — |
| Inline CSS | — |
| Inline JS | — |
| Title | — |
The same checks run over a whole site as a paid audit, plus two this tool cannot do from inside your tab: every internal link that goes nowhere, and what your site returns for a URL that does not exist — which is how you find out your 404s are being served by a parking page.
A page in your browser is not allowed to fetch another site's HTML — that is the same-origin policy, and it is doing its job. Every checker that accepts a URL is sending that URL to its own server to fetch on your behalf.
So this one does the honest version: it runs entirely in the tab, and nothing you paste leaves your machine. There is no upload, no logging, and no sign-up, because there is no server to log anything.
It counts what the markup asks for: external scripts and stylesheets, images, preloaded fonts and iframes. It does not know about requests JavaScript makes after the page loads, so a page that scores well here can still be slow — the number is a floor, not a verdict. Weight is the HTML alone; the assets it pulls are counted as requests, not bytes, because their sizes are not knowable from the source.
The same audit runs as a command-line tool over a whole site
(infra/weigh.py), and both are checked against the same fixtures in CI, so
the number here is the number that tool reports.