Your submission was sent successfully! Close

Jump to main content
  1. Blog
  2. Article

Andreea Munteanu
on 7 June 2023

Canonical’s MLOps portfolio is growing with a new machine learning tool. Charmed MLFlow 2.1 is now available in Beta. MLFlow is a crucial component of the open-source MLOps ecosystem. The project announced it had passed 10 million monthly downloads at the end of 2022. With Charmed MLFlow users benefit from a platform where they can easily manage machine learning models and workflows.
We are looking for data scientists, machine learning engineers and AI enthusiasts to take Charmed MLFlow Beta for a drive and share their feedback with us.

What is MLFlow?

MLFlow is an open-source platform used for managing machine learning workflows.  It has four primary functions that include experiment tracking, model registry, model management and code reproducibility. Like Kubeflow, MLFlow runs on any environment, including hybrid or multi-cloud scenarios,  and on any Kubernetes.

Try Charmed MLFLow Beta today

Are you a machine learning enthusiast interested in trying MLFlow? Charmed MLFlow uses Juju, Canonical’s Operator Framework, to simplify lifecycle management and integrations. It can easily be integrated with other ML tools such as Charmed Kubeflow.  Charmed MLFlow can be:

  • Deployed on your local machine
  • Used for experiment tracking, model registry, project tracking
  • Improved with your feedback

Deploying Charmed MLFlow is easy with our guide on MicroK8s.

Contact us now

Share your feedback

Charmed MLFlow is an open-source project that is growing because of the care, time and feedback that our community gives. This beta release is no exception, so if you have any feedback or questions, please feel free to contact us.

Give us your feedback

Kubeflow vs MLFlow

Kubeflow and MLFlow are both open-source projects built for machine learning initiatives. They can run on any CNCF-compliant Kubernetes and on any cloud. Kubeflow is designed for ML at scale, handling container orchestration as well as machine learning workflows. It improves the reproducibility of experiments and it allows users to perform the entire machine-learning lifecycle within one tool. MLFlow is a lightweight tool, that allows users to perform training using Python-compatible frameworks.

Both Kubeflow and MLFlow have similar capabilities, but the main difference comes from the ease of setting them up and their end-use cases. Maciej Mazur, Principal AI/ML Field Engineer and Kimonas Sotirchos, Kubeflow Working Group lead,  will host a webinar about Kubeflow vs MLFlow. Are you interested?

Join us to learn more about:

  • Open source MLOps
  • Key differences between Kubeflow and MLFlow
  • How to choose between Kubeflow and MLFlo

Register now

Learn more about MLOps

Related posts

Andreea Munteanu
23 June 2023

Kubeflow vs MLFlow: which one to choose?

AI Article

Data scientists and machine learning engineers are often looking for tools that could ease their work. Kubeflow and MLFlow are two of the most popular open-source tools in the machine learning operations (MLOps) space. They are often considered when kickstarting a new AI/ML initiative, so comparisons between them are not surprising.  This ...

Andreea Munteanu
12 July 2023

Large language models (LLMs): what, why, how?

AI Article

Large language models (LLMs) are machine-learning models specialised in understanding natural language. They became famous once ChatGPT was widely adopted around the world, but they have applications beyond chatbots. LLMs are suitable to generate translations or content summaries. This blog will explain large language models (LLMs), inclu ...

Andreea Munteanu
2 February 2023

From model-centric to data-centric MLOps

AI Article

MLOps (short for machine learning operations) is slowly evolving into an independent approach to the machine learning lifecycle that includes all steps – from data gathering to governance and monitoring. It will become a standard as artificial intelligence is moving towards becoming part of everyday business, rather than an innovative act ...