---
title: Charmhub | Deploy Kubeflow Edge using Charmhub - The Open Operator Collection
description: Deploy the latest version of Kubeflow Edge as a Kubernetes Operator on
  any cloud.
url: https://charmhub.io/kubeflow-edge/configurations/kfp-api
---

##### 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)

# Kubeflow Edge

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

* [Kubeflow Charmers](https://charmhub.io/publisher/kubeflow-charmers "View all packages from Kubeflow Charmers")
  | bundle
* [Cloud](https://charmhub.io/?filter=cloud)

Platform:

stable 46

```
juju deploy kubeflow-edge
```

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

* [Argo Controller](https://charmhub.io/kubeflow-edge/configurations/argo-controller)
* [Kfp Api](https://charmhub.io/kubeflow-edge/configurations/kfp-api)
* [Mariadb K8S](https://charmhub.io/kubeflow-edge/configurations/mariadb-k8s)
* [Kfp Persistence](https://charmhub.io/kubeflow-edge/configurations/kfp-persistence)
* [Kfp Schedwf](https://charmhub.io/kubeflow-edge/configurations/kfp-schedwf)
* [Minio](https://charmhub.io/kubeflow-edge/configurations/minio)
* [Pytorch Operator](https://charmhub.io/kubeflow-edge/configurations/pytorch-operator)
* [Seldon Core](https://charmhub.io/kubeflow-edge/configurations/seldon-core)
* [Tfjob Operator](https://charmhub.io/kubeflow-edge/configurations/tfjob-operator)

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

* auto-update-default-version | boolean

  Default: True

  If true, the default pipeline version will be updated
  when uploading a new version of a pipeline
* cache-enabled | boolean

  Default: True

  If true, pipeline run steps can be cached instead of re-run
* cache-image | string

  Default: registry.k8s.io/busybox

  Which image to list as the backing image for a pipeline run step pulled from cache
* driver-image | string

  Default: docker.io/charmedkubeflow/kfp-driver:2.4.1-1dc7f7d

  Driver image used during a pipeline's steps.
* grpc-port | string

  Default: 8887

  GRPC port
* http-port | string

  Default: 8888

  HTTP port
* init-connection-timeout | string

  Default: 6m

  Connection timeout used when initializing clients for related services.
  The format used can be anything accepted by `time.ParseDuration`.
* k8s-service-name | string

  Default: ml-pipeline

  Name of the Kubernetes service where kfp-api will expose itself
* launcher-image | string

  Default: docker.io/charmedkubeflow/kfp-launcher:2.4.1-0484071

  Launcher image used during a pipeline's steps.
* log-archive-filename | string

  Default: main.log

  Name of log file in object storage
* log-archive-prefix | string

  Default: /artifacts

  Prefix for log file in object storage
* log-level | string

  Default: info

  Log level of api server
* object-store-bucket-name | string

  Default: mlpipeline

  Default name of object storage bucket.
* runner-sa | string

  Default: default-editor

  Default pipeline runner service account.
  Used if service account is left unspecified when creating a run