Data Residency & Regional Pricing

Regional processing — routing a request through infrastructure in a specific geography for data-residency purposes — costs more than the equivalent standard request, and the uplift is published as a flat surcharge on top of the ordinary rate rather than a separate price list.

What the uplift actually covers

Data residency isn't a different model or a different capability — it's the same model, answering from infrastructure constrained to a specific region, which is a real operational cost that shows up as a percentage added to the standard rate. It applies specifically to models released on or after a stated eligibility date, so not every model on this site's roster is even eligible for regional processing in the first place.

Why this is easy to miss in a cost model

Most cost planning starts from the standard, non-regional pricing tables, and a data-residency requirement is often added to a project's constraints after the initial budget is already built — a compliance review, a customer contract, a regulatory scope that wasn't part of the original plan. The uplift needs to be added explicitly at that point; it doesn't apply itself.

Who actually needs this

Not every deployment needs data residency, and this page isn't an argument that it should — it's a statement of what it costs for the workloads that genuinely require it, whether for regulation, contractual obligation, or internal policy. For those workloads, the uplift is simply the price of meeting a real requirement, and budgeting without it isn't a rounding error; it's pricing a different, cheaper product than the one actually being deployed.

What the table below shows

The uplift percentage, the eligibility date, and which models on this site's roster qualify for regional processing at all — checked directly rather than assumed uniform across the full model lineup.