Can I select a specific Region?

Q: Can I manually select the Region (Vietnam or Japan) for Serverless Inference?

A: Currently, the Serverless Inference package does not support manual selection of the processing Region.

To optimize cost and performance for general users, our system is designed to automatically distribute API requests flexibly between the Vietnam and Japan regions.

Please refer to the table below to understand the routing mechanism and choose the service that fits your needs:

Criteria	Serverless Inference	Dedicated Inference
Routing Mechanism	Cross-Region Auto-routing Automatically switches regions (VN ↔ JP) during overloads to ensure service continuity.	Fixed Region Fixed to the Region initially selected by the customer (e.g., Running only in VN).
Data Commitment	High Availability Prioritizes availability. Data may be processed in another region during peak times.	Data Residency Prioritizes data location. Guarantees data never leaves the selected Region.
Recommendation	For customers needing convenience and flexible costs, who accept short-term offshore data processing.	For Government and Banking sectors requiring strict compliance with Data Sovereignty laws.

If your project has specific requirements for a fixed processing region (e.g., strictly running in Vietnam for data residency compliance), please switch to the Dedicated Inference service.