Google made global serving generally available for fine-tuned Gemini 3.x models on November 2 through Gemini Enterprise Agent Platform. Supported base models include Gemini 3.5 Flash (gemini-3.5-flash), Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite), and Gemini 3.1 Flash-Lite (gemini-3.1-flash-lite). Customers can copy an existing tuned model into the global region and deploy it behind a global API endpoint.
This expands deployment options for customized models, but Google's documentation identifies important limits. The global region does not support customer-managed encryption keys or Data Residency Zones, and the tuned model's adapter files are stored in the United States. Global serving therefore comes with specific data-location and encryption-control constraints for enterprise customers.
Source · docs.cloud.google.com ↗
