Is vercel.com slow?

One Vercel surface: the vercel.com marketing homepage, about 520 KB of HTML that Vercel serves from its own edge cache. It is the page Vercel publishes on its own network, so it shows how Vercel's edge answers a cold request from each of our cities. It does not stand for your project: a site you deploy on Vercel has its own build output, caching rules and functions, and its speed depends on those choices as much as on the network. We cannot see Vercel's cache state from outside, so a slow day may be a cache miss, a deploy, or the path between our worker and the nearest edge; the numbers do not say which.

Between 27 September and 10 October 2026 the typical time ranged from 57 ms in Montreal and Mumbai to 87 ms in Amsterdam, with Tokyo at 60 ms, Singapore at 62 ms and San Francisco at 67 ms. Every city was steady from day to day: Amsterdam's daily median stayed between 78 and 97 ms apart from 123 ms on 3 October, and no other city's left 44 to 94 ms. The slow week of 30 August to 3 September, when Singapore's and Tokyo's daily medians ran from 284 to 484 ms, is well outside the 14-day window; the history chart still shows it. What remains is the tail: the slowest 5% of the daily requests took 170 ms in San Francisco and 146 ms in Amsterdam, against 91 to 105 ms in the other four cities. The controlled run of 4 October put its slowest requests elsewhere, 153 to 173 ms in Mumbai and Tokyo against 67 to 108 ms in the other four. Netlify is faster in every city, by the smallest margin in Amsterdam (73 ms against 87 ms); Fastly is faster everywhere except San Francisco.