Broadcom has announced the launch of VMware Cloud Foundation (VCF) 9.1, a new AI-ready private cloud platform designed to help enterprises deploy production AI workloads with greater efficiency, security, and cost control. The latest release strengthens VMware Cloud Foundation’s position as a unified infrastructure platform for inference, agentic AI, Kubernetes, and traditional enterprise workloads.
The announcement comes as organizations increasingly shift AI workloads to private cloud environments. According to a preview of Broadcom’s Private Cloud Outlook 2026 report, 56% of organizations are already running or planning to run production AI inferencing in private clouds, while public cloud usage for production inference has declined to 41%. Rising infrastructure costs, data sovereignty concerns, and regulatory requirements are emerging as major challenges for enterprises scaling AI initiatives.
VCF 9.1 addresses these concerns through an integrated platform that combines AI and Kubernetes-native capabilities with support for mixed compute environments spanning AMD, Intel, and NVIDIA technologies. Broadcom claims the platform can reduce server costs by up to 40% through intelligent memory tiering, lower storage total cost of ownership by up to 39% using advanced compression and deduplication, and cut Kubernetes operational costs by as much as 46%.
“VCF 9.1 is a single unified platform that addresses AI infrastructure costs, data sovereignty, and agentic AI readiness, delivering one of the most advanced infrastructures for Private AI,” said Krish Prasad, Senior Vice President and General Manager, VMware Cloud Foundation Division, Broadcom.
The platform introduces enhanced resource optimization, automated fleet operations, multi-tenant infrastructure, and support for high-speed networking technologies aimed at accelerating AI deployments while maximizing infrastructure utilization. Enterprises can run both GPU-intensive inference workloads and CPU-driven agentic AI applications on a unified infrastructure layer, reducing operational complexity and eliminating the need for separate technology stacks.
Security is another major focus of the release. VMware Cloud Foundation 9.1 incorporates zero-trust architecture, ransomware recovery capabilities, continuous compliance monitoring, live patching, and distributed threat inspection to protect AI models, training data, and production workloads. The platform also extends security controls to Kubernetes environments, helping organizations maintain compliance and operational resilience as AI adoption expands.
Broadcom said the release has received support from ecosystem partners including AMD, Intel, NVIDIA, Arista Networks, and CrowdStrike, all of whom highlighted the importance of secure, scalable, and cost-efficient infrastructure for production AI environments.
With VCF 9.1, Broadcom aims to provide enterprises with a private cloud alternative that combines AI performance, operational efficiency, security, and data sovereignty while enabling organizations to accelerate their AI transformation journeys.
