AI / ML from workstation to production

Ubuntu is the platform to power your Artificial Intelligence ambitions — from developer workstations, to racks, to clouds and to the edge with smart connected IoT.

Canonical’s Kubeflow supports the most popular tools for machine learning — starting with JupyterHub and Tensorflow — in a standardised workflow running on Kubernetes.

Try Kubeflow Enterprise training

Develop your AI Models on Ubuntu

  • Workstation
  • Rack
  • Cloud
  • IoT

Develop AI models on high-end Ubuntu workstations. Train on racks of bare-metal Kubernetes or public clouds with hardware acceleration. Deploy to edge and IoT. All on Ubuntu, delivered by Canonical.

Kubeflow — AI on Kubernetes — anywhere

Google and Canonical collaborate on Kubeflow, a standardised machine learning solution for on-premises and on-cloud training. Leveraging Ubuntu, you benefit from perfect multi-cloud portability of AI/ML workloads.

Workstation AI

Ubuntu-certified workstations from Dell, Lenovo and HP with NVIDIA, microk8s and Kubeflow

  • Accelerate data science
  • Lightest footprint
  • Laptop to workstation
  • GPU optional
  • Develop and test AI

Bare metal AI

Kubernetes on bare metal with NVIDIA GPU acceleration

  • Highest performance
  • On-premises with local data
  • Hardware recommendations
  • Fully managed options

Google Cloud AI

GKE on Ubuntu with NVIDIA GPU acceleration

  • Effectively infinite scale
  • Portable workloads
  • Fastest cloud ML

Canonical Cloud AI

Kubeflow on Kubernetes on Openstack with NVIDIA GPU acceleration

  • Maximize benefits of OpenStack
  • On-premises with local data
  • Hardware recommendations
  • Fully managed options

Kubeflow features

Kubeflow brings together all the most popular tools for machine learning, starting with JupyterHub and Tensorflow, in a standardised workflow running on Kubernetes. Optimised on a wide range of hardware and cloud infrastructure, Kubeflow lets your data scientists focus on the pieces that matter to the business.

It is an extensible framework, which allows you to leverage the tools of your choice. Start with Tensorflow and JupyterHub or bring your own frameworks and tools. Combined with Kubeflow’s automation, this will accelerate your machine learning activities — from model development to model training to model sharing.

Initiated by Google on Ubuntu for perfect portability of AI workloads from your workstation, to your data center rack on Canonical’s bare-metal k8s or Canonical’s OpenStack virtualization, to Google’s Cloud Kubernetes service GKE which also runs on Ubuntu. Simple.

Canonical’s Kubeflow and Kubernetes on bare-metal 乐天堂fun88备用网址s, with NVIDIA GPUs, provides an ultra high-performance machine learning cluster. Deployment, support, and optional remote management and remote operations make it the best way to accelerate your data science and machine learning.

Canonical has provided both a familiar and highly performant operating system that works everywhere. Whether on-premises or in the cloud, software engineers and data scientists can use tools they are already familiar with, such as Ubuntu, Kubernetes and Kubeflow, and greatly accelerate their ability to deliver value for their customers.

David Aronchick, Google Product Manager for Kubeflow

in partnership with

Consulting to get started

Turn on the taps with a workshop to understand the full stack of machine learning. Build a full pipeline from developer stations to your data center, to the public cloud. Canonical works with the leading companies to ensure you have the widest range of choices. First, start with one of our standard bare-metal Kubernetes service packages (Discoverer or Discoverer Plus) and then select the AI Add-on to unlock the benefits of AI on Kubernetes.

Learn more about our Kubernetes packages ›

Kubeflow Machine Learning Starter

AI/ML pipelines and workflows on Kubernetes

Data science add-on to K8s Discoverer or Discoverer Plus. Workshop and readiness assessment covering machine learning using Kubeflow on Kubernetes for model training and analytics. Includes GPU and FPGA integration for hardware data science acceleration on k8s.


One week workshop dedicated to Kubeflow, including JupyterHub covering everything your business needs for on-prem/off-prem AI/ML operations.

  • On-site or remote options
  • Hands-on Kubernetes and Kubeflow training
  • Framework of choice: TensorFlow, PyTorch, Pachyderm, Seldon Core
  • Full pipeline view


Determine the readiness of your existing data science approach and capabilities.

  • Understand AI lifecycle
  • Preliminary data and process discovery
  • Development capacity assessment
  • Deploy and operate ML analysis
  • Finalise initial AI strategy

Get in touch

IoT and Edge AI

Train in the cloud. Act at the edge.

Cameras, music systems, cars, even firewalls and CPE are becoming smarter. From natural language processing to image recognition, from real-time high-speed autonomous navigation to network intrusion detection. Ubuntu gives you a seamless operational framework for development, training and inference all the way out to the edge.

Find out more via our dedicated webinar and blog series; parts 1, 2, 3 and 4.

Leaders in artificial intelligence choose Ubuntu

NVIDIA and Canonical accelerate AI everywhere

NVIDIA and Canonical collaborate to ensure that AI hardware acceleration is available on every public cloud, on premise and in IoT devices.

  • NVIDIA DGX 乐天堂fun88备用网址s optimised for machine learning come with Ubuntu and support from Canonical included
  • NVIDIA Kubernetes extensions for hardware acceleration are enabled in Charmed Kubernetes
  • NVIDIA Tegra and Drive PX2 ship with Ubuntu for Edge AI

Organizations are increasingly looking to accelerate their deep learning and AI implementations. In addition to using Ubuntu on our DGX systems, we have been working with Canonical to offer Kubernetes on NVIDIA GPUs as a scalable and portable solution for multi-cloud deep learning training and inference workloads.

Duncan Poole, Director of Platform Alliances at NVIDIA

Partner with us

It takes an open ecosystem to solve the diverse challenges of AI infrastructure across every sector and in every region. Our partners ensure that you have the widest range of capabilities available for automated integration in your cloud, and that you can get insight and support locally.

To learn more about our partners or becoming a Canonical AI partner, please contact us today.

Get in touch or learn more about partnering with us

Learn more about AI/ML and Kubeflow


A detailed look into the AI and ML landscape, how to deploy your first model and more.


Articles from across the web on getting started with AI and Kubeflow in your workplace.


Examine the fundamentals of a successful AI project that helps your organisation achieve their AI ambitions.

Get started with enterprise AI ›

Get started with AI and Machine learning today.

Test drive Kubeflow now

Or contact our experts to get started with consulting, training or outsourced operations.