Vercel’s AI Gateway can now pin inference to the US or the EU with a single field

A 27 July 2026 changelog entry adds regional inference to Vercel’s AI Gateway: one field pins a request to US or EU processing and works across every provider the gateway fronts, instead of per-provider configuration. The same day’s entries added WebSocket support for OpenAI’s Responses API (persistent connections and realtime voice sessions), Kimi K3 and K3 Fast with zero data retention from US-based providers, and a path to run Claude managed agents from the Chat SDK.

What it means Data residency is usually the reason an AI feature stalls in legal review; a provider-agnostic region flag turns that from an architecture problem into a configuration line.