Broadcom's VM API Project Enters CNCF Sandbox, Bridging Kubernetes and Virtual Machines

Broadcom announced today at KubeCon North America 2026 that its VM API project has been accepted into the Cloud Native Computing Foundation (CNCF) Sandbox. This project aims to enable the provisioning and management of virtual machines (VMs) through Kubernetes APIs across diverse infrastructure providers. Broadcom also revealed its co-chairmanship of the newly formed Workload Conformance Working Group within the Kubernetes community, which will focus on establishing a shared standard for how workloads run on Kubernetes. This development is significant for any organization grappling with hybrid infrastructure. While Kubernetes has become the de facto standard for container orchestration, virtual machines remain a foundational component for many enterprise applications. A recent Linux Foundation report indicates that VMs still account for 53% of production applications, a figure expected to remain stable, with AI and agentic workloads contributing to 34% of VM expansion. The VM API project directly tackles the operational overhead and complexity that arises from managing these disparate environments separately. By bringing VM management under the Kubernetes umbrella, it offers a path toward a more unified and streamlined operational model, reducing the need for specialized tools and skill sets for each technology. This move aligns with the broader trend in cloud-native computing towards abstraction and standardization. The industry has consistently sought to simplify complex infrastructure management, moving from bare metal to VMs, and then to containers orchestrated by Kubernetes. The challenge has always been to integrate legacy systems and newer paradigms seamlessly. The VM API project, by offering a vendor-neutral, community-driven approach to VM management via Kubernetes, is a natural evolution of this trend. It builds on the success of Kubernetes as a control plane for declarative infrastructure, extending its reach to traditional virtualization. This initiative also complements efforts to standardize AI workloads on Kubernetes, as many advanced AI applications still rely on the hardware isolation provided by VMs. For practitioners, this means keeping a close eye on the VM API project's progression within the CNCF Sandbox. Active participation in the Workload Conformance Working Group could be beneficial for shaping future standards. The immediate implication is the potential for simplified management of mixed workloads, reducing the learning curve and operational burden associated with maintaining separate VM and Kubernetes environments. It also offers a pathway to mitigate vendor lock-in by fostering open standards. Organizations should evaluate how this project could integrate with their existing infrastructure and consider contributing to its development to ensure it meets their specific needs, especially as AI and agentic workloads continue to drive the demand for robust, integrated infrastructure solutions.