Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

I would like to assign a custom domain to Vertex AI Endpoints.

When I deploy a model to Vertex AI Endpoints, it is assigned a default domain ({region}-aiplatform.googleapis.com). How do I switch from the default domain to the custom domain?

Solved Solved
1 2 1,194
1 ACCEPTED SOLUTION

Good day @YoshinaoMori,

Welcome back to Google Cloud Community!

Unfortunately, this is not yet possible when you are creating an endpoint in vertex ai. However if you want this feature you can submit a request using this link: https://cloud.google.com/support/docs/issue-trackers
If you are really looking for a way to assign a custom domain to the endpoint URL, you can try applying a reverse proxy that will accept the request from the client then it will forward the request to the endpoint URL.

But if you want to secure a connection for Vertex AI prediction service, It is possible to create a private endpoint for prediction, the prediction URI will look a little bit different from the Vertex AI public endpoint. Here is the format:

 

http://ENDPOINT_ID.aiplatform.googleapis.com/v1/models/DEPLOYED_MODEL_ID:predict

 

You can use this link as a guide on how to create a private endpoint: https://cloud.google.com/vertex-ai/docs/predictions/using-private-endpoints

Hope this is useful!

View solution in original post

2 REPLIES 2

Good day @YoshinaoMori,

Welcome back to Google Cloud Community!

Unfortunately, this is not yet possible when you are creating an endpoint in vertex ai. However if you want this feature you can submit a request using this link: https://cloud.google.com/support/docs/issue-trackers
If you are really looking for a way to assign a custom domain to the endpoint URL, you can try applying a reverse proxy that will accept the request from the client then it will forward the request to the endpoint URL.

But if you want to secure a connection for Vertex AI prediction service, It is possible to create a private endpoint for prediction, the prediction URI will look a little bit different from the Vertex AI public endpoint. Here is the format:

 

http://ENDPOINT_ID.aiplatform.googleapis.com/v1/models/DEPLOYED_MODEL_ID:predict

 

You can use this link as a guide on how to create a private endpoint: https://cloud.google.com/vertex-ai/docs/predictions/using-private-endpoints

Hope this is useful!

@kvandres 

Thanks for the reply!

> Unfortunately, this is not yet possible when you are creating an endpoint in vertex ai.

I understand. I will consider if it is needed and make a request!