Dataiku assets

November 6, 2026

This table shows the Dataiku objects that are integrated and how they map to Collibra asset types. The asset types are available out of the box. To use the Dataiku integration, you need Collibra Platform 2026.11 or newer and Edge version 2026.10.35 or newer.

Dataiku object	Collibra asset type	Description
A managed prediction model saved in a Dataiku project	AI Base Model	Represents foundational business concepts that define the model’s purpose and governing principles in your Dataiku instance. Only saved models of type `PREDICTION` that are managed by Dataiku are integrated. Other model types are not integrated.
A version of a managed prediction model, or an LLM Mesh model	AI Model Version	A specific, immutable implementation of the model logic and parameters that serves as the audited "raw" model version before it is activated.
Deployed API service	AI Model Deployment	The operational instance where a model version is assigned computational resources to execute and generate real-time outputs.
Deployed API service	AI Endpoint	The access point for external systems that provides a consistent interface while allowing for the seamless exchange of underlying deployments.
An agent built in Dataiku	AI Agent	An autonomous system that perceives its environment, makes decisions based on predefined rules or learned patterns, and takes actions to achieve specific goals.
A version of a Dataiku agent	AI Agent Version	A specific, immutable implementation of the model logic and parameters that serves as the audited "raw" model version before it is activated.
A tool an agent can call or a knowledge bank used for retrieval	AI Agent Tool	A service accessible to an AI agent that automates tasks, generates content, or delivers intelligent insights.
Training dataset in S3 storage (bucket)	Storage Container	Represents the S3 bucket that holds the dataset a model was trained on. To ingest it, select the Include Input Datasets option in the integration settings.
Training dataset in S3 storage (file path)	File	Represents the file or path in the S3 bucket for the training dataset. To ingest it, select the Include Input Datasets option in the integration settings.
Training dataset in a SQL table (database)	Database	A collection of data that is systematically organized or structured, to make it is easy to create, update and query the information. To ingest it, select the Include Input Datasets option in the integration settings.
Training dataset in a SQL table (schema)	Schema	An organized structure described in a formal language supported by implementing technology that defines the objects in the technology assets (Table and columns in a relational database, fields in a file). To ingest it, select the Include Input Datasets option in the integration settings.
Training dataset in a SQL table	Table	An implementation of data entities in columns and rows, in a given database system. It is the basic structure of a relational database. To ingest it, select the Include Input Datasets option in the integration settings.