Glacro takes a Git repository or an uploaded folder, runs your build on a managed runner, and serves the files that build produces over HTTPS. That is the whole product. It is worth being precise about it, because the boundary between 'static files' and 'a running application' is where most hosting disappointment lives.
What runs, and where
Builds execute on our managed build runners. The output is uploaded to our object storage, and requests to your project's URL are served from there. Everything runs in our Mumbai region. There is one region. There is no CDN in front of it and no points of presence in other countries, so a visitor in Berlin is talking to Mumbai and the latency reflects that.
Why say this out loud
The honest version costs us some sign-ups. The alternative costs somebody a migration: they build against a capability the marketing implied, deploy, and find out at the worst possible moment. We would rather be the boring option that does what it says.
What is genuinely good here
Static sites are the majority of what people deploy, and for those the trade is a good one: builds are cheap, there is no server to fall over, rollout is atomic, and billing is in rupees with GST invoices rather than a dollar amount that moves with the exchange rate. If that is your workload, this is a reasonable place to put it.