cd /news/artificial-intelligence/nutanix-enterprise-ai-brings-ai-agen… · home topics artificial-intelligence article
[ARTICLE · art-122441] src=itdaily.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Nutanix Enterprise AI brings AI agents to your infrastructure

Nutanix has launched Nutanix Enterprise AI (NAI) 2.8 and announced Nutanix Kubernetes Platform (NKP) 2.19, aiming to let organizations run AI agents and traditional applications on the same infrastructure without moving sensitive data. NAI 2.8 introduces an Agent Gateway based on the Model Context Protocol (MCP) and supports speculative decoding for up to 2.5x faster inference, while NKP 2.19 adds bare-metal Kubernetes management and an application catalog for tools like Kubeflow and Milvus.

read2 min views1 publishedSep 7, 2026
Nutanix Enterprise AI brings AI agents to your infrastructure
Image: Itdaily (auto-discovered)

Nutanix launches Enterprise AI 2.8 and announces the arrival of Kubernetes Platform 2.19. With these updates, the company aims to help organizations combine AI agents and traditional applications on the same infrastructure without the need to move sensitive corporate data.

Nutanix is making Nutanix Enterprise AI (NAI) 2.8 available immediately and simultaneously announcing the release of Nutanix Kubernetes Platform (NKP) 2.19. Many companies encounter practical barriers in AI projects. Modern AI applications almost always run in containers, while existing mission-critical applications and data are usually located on virtualized infrastructure. This creates separate IT silos with additional management overhead.

To prevent this fragmentation, Nutanix is focusing on an architecture in which virtual machines and containers function side by side. Organizations can flexibly allocate their computing power and bring AI directly to their existing data. As a result, IT teams do not need to perform complex migrations to get started with AI models.

Security and inference for AI agents #

Within NAI 2.8, Nutanix is introducing the Agent Gateway, based on the Model Context Protocol (MCP). This gateway provides AI agents with secure access to internal data and applications through a single central access point. Additionally, a new MCP server provides integration with the Nutanix Cloud Platform, allowing organizations to automate infrastructure tasks.

Inference capabilities have also been upgraded to utilize local models more efficiently across one or more graphics cards. Thanks to speculative decoding, models process requests up to two and a half times faster without loss of precision. To curb risks surrounding unauthorized software, NAI provides strict identity management and support for shielded deployments of Nvidia NIM.

Kubernetes as a foundation for AI workloads #

The Nutanix Kubernetes Platform (NKP) 2.19, which will be available soon, focuses on managing containers on bare-metal and virtual machines. With NKP Metal, the platform automates the deployment of firmware, operating systems, and storage. In combination with AHV and Nutanix Flow, the system aims to provide better network-level isolation to prevent lateral attacks.

Additionally, the platform includes an application catalog that allows teams to install software such as Kubeflow and Milvus with a single click. Partners such as Cisco and Intel provide support for the underlying computing power and hardware-based security. NAI 2.8 is available for immediate use, while the rollout of NKP 2.19 is expected shortly.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @nutanix 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/nutanix-enterprise-a…] indexed:0 read:2min 2026-09-07 ·