Hello,
This question concerns Gemini model API endpoints on Google Cloud’s Gemini Enterprise Agent Platform (formerly Vertex AI), not the Gemini Enterprise end-user application.
We currently use gemini-2.5-flash through the Vertex AI northamerica-northeast1 endpoint for a privacy-sensitive family wellness application.
We received Google’s notice that Gemini 2.5 Flash will enter Extended Lifecycle Access in October 2026. We want to migrate before the deadline, but the recommended Gemini 3.x replacement models appear to support only global, us, and eu processing locations.
Our Canadian data plane requires Gemini ML processing to remain in Canada. Using the global, US, or EU endpoint is not an acceptable migration path.
Could a Google Cloud or Vertex AI representative clarify:
- Which GA model will replace
gemini-2.5-flashinnorthamerica-northeast1? - Will Gemini 3.5 Flash-Lite—or another Gemini 3.x model—become available for Canadian ML processing before the retirement date?
- Will Gemini 2.5 Flash remain available in
northamerica-northeast1throughout Extended Lifecycle Access? - If no Gemini 3.x Canadian endpoint is planned, what migration path should customers with Canadian data-residency requirements follow?
- The email specifies October 20, 2026, while the current lifecycle documentation appears to specify October 16. Which date should customers use?
We can migrate our US workloads independently to the us multi-region endpoint. This question concerns only the Canadian workload.
Thank you.