Cannot access gemini-3.5-flash in Big Query create model

I cannot execute

CREATE OR REPLACE MODEL `<my project id>.dvo_ocr_sandbox.dvo_flash_modelxx`
REMOTE WITH CONNECTION DEFAULT 
OPTIONS (
  endpoint = 'gemini-3.1-flash-lite'
);

in my project, the response being The endpoint is not supported yet. The same applies for gemini-3.5-flash. The same command does works for gemini-2.5-flash. I have the feeling that this has to do with availability zones. The models I can use seem to be the ones with checkboxes in the specific zones in this page Deployments and endpoints  |  Gemini Enterprise Agent Platform  |  Google Cloud Documentation. The ones I cannot are the ones that only support global.

I have tried to use a full url instead of the short model name, but the error messages were the same, i.e. the global url worked only for the 2.5 models.

Is there any path to using for example gemini-3.5-flash in BigQuery?

Hello @David_Vonka,

All of the generally available and preview Gemini models are supported.

So… it should be working fine :thinking:

Looking at BigQuery REMOTE WITH CONNECTION documentation:

The Agent Platform endpoint for the remote model to use. You can specify the name of the Agent Platform model, for example gemini-2.5-flash , or you can specify the Agent Platform model’s endpoint URL, for example https://europe-west6-aiplatform.googleapis.com/v1/projects/myproject/locations/europe-west6/publishers/google/models/gemini-2.5-flash . If you specify the model name, BigQuery ML automatically identifies and uses the full endpoint of the Agent Platform model based on the location of the dataset in which you create the model.

I think that your intuition about availability zones is right.

The problem is that even if I go with locations/global/publishers… it still does not work