ML metadata artifact types

For the purpose of tracking resources and for functionality across the Google Cloud Pipeline Components, a set of standard Vertex ML Metadata artifact types are defined. These first-party artifacts types are defined by Google Cloud Pipeline Components.

Artifact type	Description
`google.VertexModel`	An artifact representing a Vertex AI `Model` resource
`google.VertexEndpoint`	An artifact representing a Vertex AI `Endpoint` resource
`google.VertexBatchPredictionJob`	An artifact representing a Vertex AI `BatchPredictionJob` resource
`google.VertexDataset`	An artifact representing a Vertex AI `Dataset` resource
`google.BQMLModel`	An artifact representing a BigQuery ML `Model` resource
`google.BQTable`	An artifact representing a BigQuery `Table` resource
`google.UnmanagedContainerModel`	An artifact representing an `UnmanagedContainerModel` resource

For more information about using ML metadata artifacts, see Consume or produce artifacts in your component.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2025-10-31 UTC.