How we measured
We created a site with npm create vagejs@latest my-site, built it with npx vage build and served it with npx vage preview. Then we ran Lighthouse 13.5.0 on the home page with its default mobile settings: a simulated slow 4G connection (150 ms round trips, about 1.6 Mbit/s) and a CPU slowed down four times. We ran it twice on 28 October 2026 and got the same results both times.
Results
| Measure | Result | Good is |
|---|---|---|
| Performance | 100 | 90 or more |
| Accessibility | 100 | 90 or more |
| Best Practices | 100 | 90 or more |
| SEO | 100 | 90 or more |
| Agentic Browsing | 100 | 90 or more |
| Main content appears (LCP) | 0.8 s | under 2.5 s |
| First text appears (FCP) | 0.6 s | lower is better |
| Layout jumps (CLS) | 0 | under 0.1 |
| Time the page is frozen (TBT) | 0 ms | lower is better |
| Data downloaded | 1.7 KB in 2 requests | less is faster |
| JavaScript sent | 0 KB | less is faster |
The full report
Download the Lighthouse report (JSON, 333 KB). To read it as the usual Lighthouse page, open the Lighthouse Report Viewer from the Lighthouse documentation and drop the file onto it, or load it in Chrome DevTools.
What these numbers don't tell you
These are lab results for the starter site on one computer. Your own site's numbers depend on your images, fonts, content and host, and Google also looks at data from real visitors (Core Web Vitals in the Chrome UX Report), which a new site doesn't have yet. Run npx vage audit --lighthouse on your own site to measure it the same way.