MLflow is a machine learning (ML) platform that helps you track, store, and manage the metadata of your ML models and AI agents. When you integrate MLflow, you integrate the metadata of your ML models and AI agents from MLflow to Collibra Platform.
Agent detection requires MLflow 3. MLflow 2 registered models are always ingested as AI Base Model and MLflow AI Model Version assets. When you synchronize a registered MLflow model, the MLflow AI integration checks whether it is backed by an AI agent. If available, the integration ingests it as an AI Agent and AI Agent Version asset instead of as an AI Base Model and MLflow AI Model Version asset. A registered MLflow model always produces one of these two asset pairs. If the integration cannot confirm that a registered model is an agent, it falls back to the standard AI Base Model and MLflow AI Model Version ingestion path.
-
You can configure Edge connections and capabilities without an active AI Governance license. However, AI Governance must be enabled to harvest AI model metadata, ingest corresponding AI assets in Data Catalog, and access the dashboards and features necessary to visualize and govern your AI landscape.
- You can integrate MLflow only via Edge.
For information about MLflow, go to the MLflow documentation.