Nutanix (NASDAQ: NTNX) announced the launch of its Nutanix Agentic AI solution at NVIDIA GTC, a software stack designed to help enterprises deploy and manage AI agent applications at scale.

The solution integrates with NVIDIA AI Enterprise and includes infrastructure orchestration, security software, AI Platform Services, and Models-as-a-Service capabilities. It aims to address the complexity of managing infrastructure required to run thousands of AI agents simultaneously in enterprise environments.

The Nutanix Agentic AI solution consists of three main components: AI services and Kubernetes platform, infrastructure optimization and security features, and foundational data services. The latest version of Nutanix Enterprise AI (version 2.6) includes an AI Gateway service for policy control over cloud-hosted and private large language models, plus support for the Model Context Protocol server and Fine Tuning capabilities.

The solution extends Nutanix's AHV hypervisor with topology-aware capabilities to optimize resource allocation on GPU-dense servers. The Flow Virtual Networking solution has been enhanced to offload network dataplane processing to NVIDIA BlueField, aiming to reduce host CPU and memory consumption.

"Production Agentic AI infrastructure needs to handle scale and high rates of change for thousands of AI services, agents, and concurrent users and developers," said Thomas Cornely, Executive Vice President of Product Management at Nutanix.

The solution can be deployed on hardware from Cisco, Dell, and Supermicro with joint validation by Nutanix and NVIDIA. Partner companies including TCS, Accenture, and Fujitsu have announced support for the platform.

Components of the Nutanix Agentic AI solution are either generally available or in early access, with broader availability expected soon.