Predict
vertex_ai.predict
ActionGoogle Cloudconnection required
Calls the Vertex AI endpoints/{endpoint}:predict endpoint.
Send instances to a deployed Vertex AI endpoint and return its predictions.
Connection
- Required
- Yes — the node will not save without one
- Auth type
- OAuth 2.0
- Providers
- Google Vertex AI
- Secret shape
- {"client_id", "client_secret", "refresh_token"}
- Config it needs
- project_id, region
- Resolved by
- config.connection_id, else node.connection_id
Configuration
5 fields, 2 required. A dotted path is a nested object and name[] an array of objects — the editor renders the same tree.
| Field | Type | Required | Default | Constraints | Example |
|---|---|---|---|---|---|
project_id GCP project id (overrides connection). | string | No | — | — | — |
region Region (default us-central1). | string | No | — | — | — |
endpoint_id Endpoint id (numeric). | string | Yes | — | — | — |
instances Instances (JSON array). | array | Yes | — | ≥ 1 characters; array of object | — |
parameters Prediction parameters (JSON). | object | No | — | — | — |
Output
One shape.
result
| Field | Type | Example |
|---|---|---|
predictions | array | — |
count | number | — |
deployed_model_id | string | — |
{
"count": 0,
"deployed_model_id": "",
"predictions": []
}Related Google Vertex AI nodes
See Predict run in a workflow.
In a technical demo we add the Google Vertex AI connection, drop the node on the canvas and let Workflow Health pre-flight it before the first real run.