Microsoft Foundry adds region-agnostic reservations for
Global Provisioned Throughput Units (PTUs) at general availability, letting
customers commit to PTU capacity once and consume it from any supported region.
Previously, a reserved PTU commitment was tied to a single region, which
created friction for customers that operated multi-region applications or
wanted to shift workloads as traffic patterns changed. With region-agnostic
reservations, finance teams can sign one reservation contract and engineering
teams can deploy models in whichever Azure region best fits a given workload,
with the reservation discount applied automatically. Reservations cover
supported Foundry Models including Azure Direct first-party models such as
GPT-5 and GPT-4.1. Existing region-bound reservations continue to honor their
original terms.
Learn more.