CODEWITHSUNDEEP
google-cloud-platformgoogle-cloud-vertex-ai
kunwarvikrant
kunwarvikrant

Reputation: 53

How do you make vertex ai prediction endpoint access public with simple authentication?

I have created a custom container model on the Google Vertex Ai platform to deploy a model, I have also deployed it to an endpoint,I can access it using bearer token and OAuth 2.0 set up but i was wondering if there is a way to make this prediction endpoint public for testing with a simple auth key or something?

Upvotes: 3

Views: 2297

Answers (1)

Chris
Chris

Reputation: 1455

I was wondering if there is a way to make this prediction endpoint public for testing with a simple auth key?

Yes, you can use a customer managed encryption key (CMEK) with no rotation policy as a simple symmetric authentication key. This guide tells you how.

Once the key is imported, choose the CMEK option and select your key under the Encryption section when creating the Vertex AI endpoint.

In addition, choosing "Standard Access" when creating the endpoint creates a publicly reachable endpoint.

(Initially I got an error that quota metrics exceed quota limits for in_use_customer_managed_encryption_keys. However, I successfully applied for a quota increase and was subsequently able to create the endpoint with the CMEK key).

Upvotes: 0

Related Questions