Kubeflow, the open source project that extends machine learning (ML) stack support on Kubernetes, has graduated, having matured into a widely adopted industry standard.
The platform sits as part of the Cloud Native Computing Foundation (CNCF)’s Kubernetes project and debuted in late 2017. It was designed by Google engineers to bridge the ML accessibility gap for developers while expanding Kubernetes outside of its original container orchestration focus.
Today, Kubeflow Python packages have reached almost 260 million downloads counting users from Nvidia, LinkedIn, and Spotify. It joined CNCF in 2023 as an incubating project and now features more than 6,600 contributors.
The rise of AI has seen Kubeflow’s relevancy grow, with it being used to aid enterprises across data processing, model training, and fine-tuning. And it’s set to scale further with the project’s roadmap focusing on expanding large language model (LLM) orchestration and enhancing post-training capabilities.
Beyond AI, Kubeflow has helped with other CNCF technologies, including Prometheus for monitoring and Istio for secure service communication.
Graduation for an open source project only occurs once it’s mature enough to become widely adopted. Exact requirements see projects subjected to a third-party security audit and a formal steering committee put in place to take charge of governance. The CNCF Code of Conduct must also be adopted to achieve graduation status.
“Nine years ago, Jeremy Lewi, Vishnu Kannan and I put together a crazy demo involving hot dogs and Kubernetes, and Kubeflow was born,” said Kubeflow co-founder and Expanso CEO David Aronchick. “I could not be more ecstatic to see how far it’s come – and how many people have turned it into something teams and businesses genuinely rely on.”
“Kubeflow has become a fantastic platform for organizations looking to unify work across AI, data science, and platform engineering teams,” said CNCF CTO Chris Aniszczyk. “Graduation marks a critical milestone, cementing Kubeflow as a mature option for enterprise AI workloads on Kubernetes. The project’s remarkable growth reflects the tireless work of its maintainers and community, and we are thrilled to celebrate this milestone with them.”