We are proud to announce OCI Kubernetes Engine (OKE) is part of the first group of Kubernetes providers to be certified under the Cloud Native Computing Foundation‘s (CNCF) Kubernetes AI Conformance Program. This certification signals a new standard for AI/ML workloads running on Kubernetes.
Why this matters
As more AI/ML workloads move into production, practitioners and the organizations that support them need a consistent, reliable foundation to build on. The CNCF previously launched the Certified Kubernetes Conformance Program, which established a set of required APIs for Kubernetes providers to support in order to ensure interoperability from one Kubernetes installation to the next, as well as compatibility with open source, third party tools. Launched at KubeCon + CloudNativeCon North America 2025, the CNCF’s Kubernetes AI Conformance Program applied this same model to AI/ML infrastructure. It establishes a minimum set of capabilities, APIs, and configurations required to reliably and efficiently run AI/ML workloads on Kubernetes. These criteria are defined by the community through the Kubernetes AI Conformance Working Group.
Certification means a Kubernetes environment is qualified to handle demanding AI training and inference workloads, through support for advanced scheduling, topology awareness, GPU accelerator management, monitoring, and more. By achieving this certification, users have the confidence needed to run cloud scale AI/ML on the OKE platform.
Why OCI was well-positioned to earn this certification
“At Oracle, we see open standards as critical for sustainable and scalable AI innovation. The AI/ML space is dynamic, and our users are looking for consistency. By supporting Kubernetes AI Conformance, we’re helping to reduce fragmentation of architectures and technologies to ensure developers and enterprises have a foundation to run production-ready AI/ML workloads reliably and efficiently.”
Sudha Raghavan, senior vice president, AI Infrastructure, Oracle Cloud Infrastructure and CNCF board member
From day one, OCI Kubernetes Engine (OKE) was built with cloud native best practices and extensibility in mind. We have invested in providing an open Kubernetes platform that enables our users to take advantage of the high performance compute, networking, storage and more offered by Oracle Cloud Infrastructure, including the latest shapes from Nvidia and AMD. OKE’s participation in the cloud native ecosystem combined with Oracle Cloud Infrastructure’s focus on GPU infrastructure, made it possible to meet these rigorous criteria of the Kubernetes AI Conformance program. For more information regarding AI/ML workloads running on OCI, check out the Kubecon North America 2025 keynote from Cohere: Cloud Scale Enterprise AI: How Cohere Runs on Open Source with Oracle Cloud and OCI AI/ML customer success stories.
Next steps
If you’re running or planning AI/ML workloads on Kubernetes, we invite you to explore OKE to experience the scale, performance, and reliability that come with a certified platform. Talk to your Oracle account team about how to get started with an AI-ready Kubernetes deployment.
Thank you to all the OCI engineering and product teams who made this possible and to our customers and partners who trust us to deliver infrastructure that helps power tomorrow’s AI innovations.
For more information on Kubernetes AI conformance, read the CNCF Launches Certified Kubernetes AI Conformance Program to Standardize AI Workloads on Kubernetes launch announcement and check out the Certified AI Platform Conformance Program project page.