Skip to main content
GPU infrastructure across five regions on three continents. Requests route to the nearest region by default.

Current Regions

RegionLocationData CenterWorkloads
US EastNew Jerseynj1AI Assistants, Voice AI
EU CentralFrankfurtde1AI Assistants, Voice AI
EU WestParispa1AI Assistants, Voice AI
EU WestAmsterdamat1AI Assistants, Voice AI
Asia-PacificSydneysy1AI Assistants, Voice AI

Routing

Requests route automatically based on your account’s data locality setting. No region parameter needed. To pin traffic to a specific region for compliance, contact support.

Data Residency

All inference processing runs in-region. For EU accounts, requests and responses stay within the EU.

Roadmap

  • Region selection API parameter
  • Per-region model status and latency metrics
  • Edge inference for sub-50ms response times