#Introduction

Alauda support for Kubeflow provides a Kubernetes-native machine learning platform that enables users to build, deploy, and manage machine learning models at scale. It integrates with various components of the Kubeflow ecosystem, such as Kubeflow Pipelines for workflow orchestration, Kubeflow Trainer v2 for training job management, and Kubeflow Model Registry for model versioning and management.

Starting in v26.3.0 (Alauda AI v2.3), Alauda Kubeflow components are shipped as OLM Helm Operators installed from the ACP OperatorHub: kfbase-operator, kfp-operator, and kubeflow-trainer-operator. The earlier Cluster Plugin form factor (v1.x) is retired; see Upgrade Kubeflow Operators for the migration path. The aligned upstream is kubeflow/manifests release 26.03.

See Kubeflow Docs for more details about Kubeflow components and features.

**NOTE: ** You need to set the namespace PSA to privileged in order to use Kubeflow components. Please contact your cluster administrator to set the namespace PSA to privileged if you encounter permission issues when using Kubeflow components.