1. Undeploy the model as follows:

   1. In the Google Cloud console, in the Vertex AI section, goto theEndpoints page.

      Go to the Endpoints page

   2. Click the name and version ID of the model you want to undeploy to openits details page.

   3. On the row for your model, clickmore_vertActions, and then clickUndeploy model from endpoint.

   4. In theUndeploy model from endpoint dialog, clickUndeploy.

   5. To delete additional models, repeat the preceding steps.

2. Optional: Delete the online inference endpoint as follows:

   1. In the Google Cloud console, in theVertex AI section,go to theOnline prediction page.

      Go to Online prediction

   2. Select the endpoint.

   3. To delete the endpoint, clickmore_vertActions, and then clickDelete endpoint.

gcloud

1. List the endpoint IDs for all endpoints in your project:

   gcloudaiendpointslist\--project=PROJECT_ID\--region=LOCATION_ID

   ReplacePROJECT_ID with your project name andLOCATION_ID with the region where you are usingVertex AI.

2. List the model IDs for the models that are deployed to an endpoint:

   gcloudaiendpointsdescribeENDPOINT_ID\--project=PROJECT_ID\--region=LOCATION_ID

   ReplaceENDPOINT_ID with the endpoint ID.

3. Undeploy a model from the endpoint:

   gcloudaiendpointsundeploy-modelENDPOINT_ID\--project=PROJECT_ID\--region=LOCATION_ID\--deployed-model-id=DEPLOYED_MODEL_ID

   ReplaceDEPLOYED_MODEL_ID with the model ID.

4. Optional: Delete the online inference endpoint:

   gcloudaiendpointsdeleteENDPOINT_ID\--project=PROJECT_ID\--region=LOCATION_ID