Vertex AI Endpoint

gcp-vertex-ai

A Vertex AI Endpoint — the regional serving surface you deploy trained models behind for online prediction. The endpoint is a free control-plane object; cost only begins when you deploy a model (the replica VMs). This module provisions the endpoint with optional CMEK, optional private (Private Service Access) networking and optional request/response logging, and leaves model deployment to you — so it stands up and tears down for cents. Works with Terraform and OpenTofu (>= 1.6), Google provider >= 7.0, < 8.0.

Defaults:

• Public endpoint (network = null) reachable over the Google API surface
• Google-managed encryption (kms_key_name = null)
• No request/response logging unless you opt in

This module deliberately creates no VMs, notebooks or Workbench instances (those are billed compute). To actually serve predictions, deploy a model to the endpoint separately (google_vertex_ai_endpoint_with_model_garden_deployment or gcloud ai endpoints deploy-model).

Requirements

• Terraform or OpenTofu >= 1.6
• hashicorp/google >= 7.0, < 8.0
• API: aiplatform.googleapis.com

Verification

Static-validated (fmt, validate, tflint). Live apply/destroy testing pending cloud sandbox availability — see catalog status.

License

Commercial — IaC Bazaar EULA. © IaC Bazaar. Original work (not derived from a third-party module).