Placement reflects growing enterprise demand for open, portable infrastructure across Kubernetes, cloud and AI
Mirantis, delivering Kubernetes-native infrastructure for AI, announced that it has been named a Challenger in the 2026 Gartner® Magic Quadrant™ for Container Management.
Also Read: CIO Influence Interview with John Elliott, Cybersecurity Author Fellow at Pluralsight
The recognition comes as enterprises increasingly need to manage Kubernetes and containerized workloads that span bare metal, virtualization, public clouds and the edge. At the same time, the rapid growth of AI is adding GPU infrastructure to that mix, increasing the need for a consistent operating model that gives organizations control over where and how workloads run.
“We believe being named a Challenger by Gartner validates the direction we have taken with Mirantis: giving enterprises a consistent, open way to operate increasingly complex infrastructure without locking themselves into a single cloud, virtualization platform or proprietary ecosystem,” said Alex Freedland, CEO of Mirantis. “That becomes even more important as organizations build AI infrastructure. Platform teams need holistic management of Kubernetes, virtual machines, GPUs and traditional infrastructure while preserving their freedom to run workloads where they make the most technical and economic sense.”
One Operating Model Across Kubernetes, Cloud and AI Infrastructure
At the center of the Mirantis strategy is k0rdent, its open source orchestration framework for managing distributed infrastructure and Kubernetes. k0rdent provides a unified control plane spanning bare metal, virtualization and public clouds, helping platform teams reduce operational silos while applying consistent governance, security policies and GitOps workflows from lightweight k0s clusters at the far edge to large-scale Mirantis Kubernetes Engine (MKE) deployments in the datacenter.
The Mirantis open source portfolio includes k0s, its lightweight, certified Kubernetes distribution, and Mirantis OpenStack for Kubernetes, which brings virtual machines and traditional infrastructure into a Kubernetes-based operational model.
With k0rdent AI, Mirantis extends that approach to GPU infrastructure and AI workloads. Rather than treating AI infrastructure as a separate technology requiring its own operational model, Mirantis enables organizations to manage GPU-backed Kubernetes clusters alongside the infrastructure supporting their existing applications.
This Metal-to-Model approach is designed to provide a consistent infrastructure foundation across the AI lifecycle, helping organizations automate operations, improve GPU utilization and scale AI infrastructure without creating another isolated infrastructure stack.
Open Infrastructure Without Proprietary Lock-In
Mirantis takes an open source and standards-based approach that is designed to preserve infrastructure portability. k0rdent and k0s are open source, and Mirantis emphasizes open APIs rather than proprietary interfaces that can make moving applications and infrastructure more difficult.
This approach gives organizations greater freedom to determine where workloads run based on performance, economics, security, sovereignty and regulatory requirements rather than the limitations of a particular infrastructure provider.
It is particularly relevant for enterprises with sovereignty requirements across hybrid cloud, multi-cloud, and AI infrastructure, as well as organizations in telecommunications, financial services and the public sector where infrastructure decisions increasingly must account for security, reliability, compliance and operational control.
“Enterprise infrastructure is moving toward a world where Kubernetes, virtualization and AI infrastructure have to coexist,” Freedland added. “Our goal is to eliminate the boundaries between those and provide organizations with one open operating model across all of them.”
Catch more CIO Insights: How Are CIOs Aligning Technology with Workforce Agility?
[To share your insights with us, please write to psen@itechseries.com ]
