Regions

A region is the physical location your project's storage and compute run in. You pick it once, when you create the project; every branch and endpoint under that project inherits it. Region is a property of the project, not something you change later.

The region catalog

These are the regions you can create projects in:

An em-dash means the catalog does not carry that field yet.

GET /v1/regions returns the same list. It is authenticated — the same bearer that lists your projects lists the catalog. The response is a single object:

{
  "regions": [
    {
      "region_id": "gcp-usc1-dev",
      "name": "gcp-usc1-dev",
      "default": true,
      "geo_lat": 0,
      "geo_long": 0
    }
  ]
}

Each region carries:

• region_id — the stable identifier you pass when creating a project.
• name — the display label. There is no display-name column yet, so it repeats region_id.
• default — whether this is the region the console preselects.
• geo_lat / geo_long — coordinates for distance-based selection, 0 until the directory carries real ones.

Choosing a region at create time

In the web console, New project has a Region dropdown. It defaults to Auto (nearest), which leaves the choice to the control plane. To pin a specific region, select it from the list.

Over the API, pass the optional region field on POST /v1/projects:

{ "name": "my-app", "region": "gcp-usc1-dev" }

Omit region (or send null) for the same "auto" behaviour as the console default. Auto is unambiguous only while exactly one region is active: with more than one, a create naming no region is rejected with region_required (422) instead of the control plane guessing.

Region availability

The table above is generated from the catalog, not hand-maintained, so it is what you can create in right now. Regions are added as demand grows. Nothing changes on the client when one comes online — the catalog and the region field are already in place.

Related

• Projects — the container a region is chosen for.
• Branches — inherit their project's region.