---
title: Charmhub | Deploy Mlflow using Charmhub - The Open Operator Collection
description: Deploy the latest version of Mlflow on any cloud.
url: https://charmhub.io/mlflow/configurations/mlflow-server
---

##### We've discontinued the registration of new Bundles

New Bundle registrations are no longer accepted. Existing bundles remain functional. We recommend using the Juju Terraform Provider for new deployments.

[Learn more](https://discourse.charmhub.io/t/discontinuing-new-charmhub-bundle-registrations/15344)

# Mlflow

[Kubeflow Charmers](https://charmhub.io/publisher/ckfbot "View all packages from Kubeflow Charmers")
| bundle

* [Kubeflow Charmers](https://charmhub.io/publisher/ckfbot "View all packages from Kubeflow Charmers")
  | bundle
* [Ai Ml](https://charmhub.io/?filter=ai-ml)

Platform:

stable 52

```
juju deploy mlflow
```

[Learn to deploy on juju >](https://juju.is/docs/juju/manage-applications)

* [Minio](https://charmhub.io/mlflow/configurations/minio)
* [Mysql K8S](https://charmhub.io/mlflow/configurations/mysql-k8s)
* [Mlflow Server](https://charmhub.io/mlflow/configurations/mlflow-server)

[Learn about configurations >](https://juju.is/docs/juju/configuration#heading--application-configuration)

* create\_default\_artifact\_root\_if\_missing | boolean

  Default: True

  If True, charm will try to create the default\_artifact\_root bucket in S3 if it does not
  exist. If False and the bucket does not exist, the charm enter Blocked status
* default\_artifact\_root | string

  Default: mlflow

  The name of the default bucket mlflow uses for artifacts, if not specified by the workflow
* enable\_mlflow\_nodeport | boolean

  Default: True

  Deploy the NodePort service for MLFlow
* mlflow\_nodeport | int

  Default: 31380

  The nodeport for MLFlow
* mlflow\_port | int

  Default: 5000

  The port MLFlow will be listening on
* mlflow\_prometheus\_exporter\_nodeport | int

  Default: 31381

  The nodeport for MLFlow Prometheus exporter
* mlflow\_prometheus\_exporter\_port | int

  Default: 8000

  The port MLFlow Prometheus exporter will be listening on