MLflow Kubeflow Integration

This repository provides the integration layer between MLflow and Kubeflow, making MLflow the first-class experiment tracking experience for the Kubeflow platform (KEP-897).

This repository packages two MLflow extensions for Kubernetes-backed deployments:

• a workspace provider that maps MLflow workspaces to Kubernetes namespaces
• an optional authorization plugin that enforces Kubernetes RBAC for MLflow requests

These plugins build on top of MLflow's workspace support. If you are new to MLflow workspaces, start with the official guide: https://mlflow.org/docs/latest/self-hosting/workspaces/getting-started/.

Components

Install

The v1 release line supports MLflow >=3.10.0,<3.17.0, including MLflow 3.16.

Install from PyPI:

pip install mlflow-kubernetes-plugins

For local development:

uv sync --extra dev

For Kubernetes deployment, see the charts/mlflow installation guide.

Quick Start

1. Enable MLflow workspaces on an MLflow server backed by a SQL store.
2. Install this package into the same environment as the MLflow server.
3. Configure the workspace provider and, if needed, the auth plugin.

export MLFLOW_K8S_WORKSPACE_LABEL_SELECTOR="mlflow-enabled=true"
export MLFLOW_K8S_DEFAULT_WORKSPACE="team-a"

mlflow server \
  --backend-store-uri postgresql://user:pass@localhost/mlflow \
  --default-artifact-root s3://mlflow-artifacts \
  --enable-workspaces \
  --workspace-store-uri "kubernetes://" \
  --app-name kubernetes-auth

Use --app-name kubernetes-auth only when you want request authorization enforced by Kubernetes RBAC.

Documentation

• docs/index.md: docs index
• docs/workspace-provider.md: workspace provider behavior, configuration, and startup
• docs/authorization-plugin.md: auth modes, headers, and request handling
• docs/kubernetes-rbac.md: RBAC requirements and example manifests
• config/crd/bases/mlflow.kubeflow.org_mlflowconfigs.yaml: generated MLflowConfig CRD manifest

Development

See CONTRIBUTING.md for development setup, coding style, and testing instructions.