Nutanix Inc. is used its .NEXT conference this week to outline a broad expansion of its cloud platform, positioning the company to support what it describes as the next phase of enterprise artificial intelligence with agent-driven applications running across hybrid and multicloud environments.

Central to today’s announcements are new capabilities in the Nutanix Cloud Platform, including infrastructure for agentic AI, expanded Kubernetes support, deeper ecosystem integrations and enhanced management tools. Nutanix said the updates reflect growing enterprise demand for platforms that can handle increasingly complex workloads while navigating hardware supply constraints and regulatory requirements.

“The theme of the show is one platform and one experience only,” said Lee Caswell, senior vice president of product and solutions marketing. “The idea is that platforms are expanding to include workloads of all types that can run anywhere.”

Nutanix is also extending a previously announced agentic AI stack with new multitenant and service provider capabilities designed to support “neoclouds,” or providers offering on-demand access to AI infrastructure and services.

The company said the platform will enable these providers to deliver services such as graphics processing units-as-a-service and Kubernetes-as-a-service, while also supporting enterprise use cases that require governance, cost control and data sovereignty. A new management layer allows providers to allocate shared AI infrastructure while maintaining isolation between tenants.

Caswell emphasized that managing cost and performance will be critical as AI workloads scale. He emphasized the importance of monitoring token usage in large language model environments to avoid unexpected expenses, describing it as “important as customers look at how they deploy AI without getting surprised about the costs.”

Bare-metal convergence

Another element of the announcement is NKP Metal, an extension of the Nutanix Kubernetes Platform that allows Kubernetes workloads to run directly on bare-metal infrastructure. The offering is aimed at performance-sensitive use cases such as AI training and edge deployments that rely on dense GPU configurations.

Running Kubernetes on bare metal usually introduces operational complexity in provisioning physical servers and managing firmware. Nutanix is attempting to address that by applying its existing automation, lifecycle management and data services to such environments.

The company is also promoting what it calls a “dual-native” architecture, enabling organizations to run containers either on virtual machines or directly on bare metal under a unified management and security model. Caswell said Nutanix is the only company that offers such an approach. It’s intended to offer organizations flexibility as they balance performance, cost and operational consistency across environments.

Expanded ecosystem

Nutanix is also expanding its ecosystem of hardware and cloud partners to include support for additional storage platforms, server vendors and cloud environments.

The company highlighted new and planned collaborations with leading vendors such as Cisco Systems Inc., Dell Technologies Inc., Lenovo Group Ltd. and NetApp Inc., as well as expanded support for Advanced Micro Devices Inc.-based systems. The integrations are intended to let organizations reuse existing infrastructure and mitigate supply chain constraints that have affected hardware availability.

The platform also adds zero-copy migration capabilities from VMware Inc. vSphere environments to Nutanix AHV virtual disks, enabling in-place workload conversion without duplicating data.

For enterprises expanding across multiple clouds and geographic regions, Nutanix is also emphasizing data sovereignty and operational control. Updates to Nutanix Cloud Clusters expand support to additional hyperscaler cloud regions, including sovereign environments. This allows organizations to meet regulatory requirements while maintaining workload portability, Nutanix said.

The company is also enhancing Nutanix Cloud Manager with multisite and multidomain capabilities. Version 2.0 provides a single control plane for managing distributed infrastructure, including air-gapped and highly regulated environments.

New features include integrated cost governance, AIOps and self-service capabilities, all delivered with a unified interface. By bringing cost management on-premises, Nutanix said it eliminates the need for customers to rely on cloud-based tools in sensitive environments.

The company is also expanding its Data Lens ransomware protection offering to provide analytics and governance capabilities in on-premises and air-gapped deployments.

Preparing for the next phase of AI

Taken together, the announcements reflect Nutanix’s effort to reposition its platform beyond its roots in hyperconverged infrastructure toward a broader role in AI and multicloud operations.

The company is betting that enterprises will increasingly seek integrated platforms that can support AI workloads alongside traditional applications, while providing consistent management, governance and cost control across environments.

Caswell framed the shift as a continuation of Nutanix’s longstanding “run anything anywhere” philosophy, extended to include AI and modern application architectures.

“We’re bringing the simplicity that Nutanix has been known for into bare metal,” he said. “This allows us to consolidate management and free up time for other, more high-value capabilities.”

Many of the new capabilities are available now. Nutanix Agentic AI,  Remote Direct Memory Access for Nutanix Unified Storage, enhanced service provider features and the NKP Metal deployment option will be available in the second half of 2026.

Photo: Flickr CC