Hybrid enterprise AI has emerged as a cornerstone for enterprises looking to balance on-premises and cloud-based deployments. As artificial intelligence shifts from a future ambition to a present-day imperative, organizations demand more accessible, adaptable and cost-effective solutions.

Answering that call, Nutanix Inc. has brought its unified hybrid strategy into sharp focus this year during industry-shaping events, including speaking with theCUBE during AWS re:Invent, Microsoft Ignite, KubeCon + CloudNativeCon NA and Nutanix’s The AI-Ready Platform event. With a vision centered on accessibility, operational simplicity, scalability and strategic partnerships, Nutanix drives the next phase of enterprise AI adoption, empowering organizations to deploy AI on their own terms.

“Nutanix’s renowned cloud operating model — spanning storage, virtualization and containers — now simplifies generative AI in a truly innovative way,” said Rob Strechay, managing director with theCUBE Research. “Building on its rich history of delivering user-friendly infrastructure, Nutanix Enterprise AI extends its capabilities beyond the Nutanix platform to hyperscale clouds, leveraging cloud-native Kubernetes for seamless generative AI infrastructure.”

This feature is part of SiliconANGLE Media’s exploration of Nutanix’s hybrid enterprise AI strategy, highlighting its efforts to simplify AI deployment across on-premises, cloud and edge environments through partnerships, operational simplicity and flexible infrastructure solutions. (* Disclosure below.)

Simplifying enterprise AI deployment for all environments

Nutanix’s debut of GPT in a Box at AWS re:Invent marked a significant milestone for the company, streamlining the deployment of AI models on Amazon Elastic Kubernetes Service with minimal technical friction. Paired with on-premises generative AI capabilities the company showcased at KubeCon, this approach supports multi-environment enterprise AI implementations, according to Luke Congdon, senior director of product management at Nutanix.

“We just released what we’re calling Nutanix Enterprise AI, and it’s our new product line to say we are an AI company,” Congdon told theCUBE during KubeCon. “We are offering generative AI on-premises now with inference endpoints, with security, with cost control, with simplicity, which is what we’ve always been trying to do.”

This solution features a tokenized cost model and a user experience similar to OpenAI’s tools, simplifying the deployment of models from Nvidia Corp. and Hugging Face Inc. By addressing common barriers to adoption, Nutanix enables companies to achieve faster and more efficient enterprise AI implementations, according to Debojyoti Dutta, vice president of engineering at Nutanix.

“The way the product works is very simple. We simplify the entire lifecycle of inference for our customer,” Dutta said in an interview during AWS re:Invent. “A customer can go and choose any model from Hugging Face or from the Nvidia catalog and then deploy the model very easily with a couple of button clicks.”

Nutanix’s focus on multi-environment support provides a consistent experience across cloud and on-premises environments, reducing operational friction. This unified approach to AI deployment ensures enterprises can navigate complex lifecycles with ease, according to Congdon.

“We’d like to make sure that you’re going to get production-level Kubernetes with security, with load balancing, with ingress, with everything else that you need, because Kubernetes is really great, but it’s one key important orchestrator piece,” Congdon said. “You need so much more. What’s really unique about them coming to Nutanix is they got what has traditionally really been hard for Kubernetes … stateful storage across objects, files, volumes, anything that you need.”

Deploying enterprise AI anywhere, from the cloud to the edge

Nutanix’s hybrid AI strategy enables enterprises to deploy AI models across on-premises, public cloud and edge environments, offering the flexibility to run workloads where they make the most sense. This unified operating model addresses diverse infrastructure setups and data residency requirements, according to Lee Caswell, senior vice president of product and solutions marketing at Nutanix.

“We brought a cloud operating model to an on-premises environment,” he said during an interview at Microsoft Ignite. “[We’ve] now extended that operating model into Azure natively so that you can go and run … the same operating model you run on-prem and all of that enterprise resilience, the day two operations and security and privacy you can now extend into Azure … extends now into AI.”

This flexibility is especially crucial for healthcare, finance and government industries, where data control and latency are non-negotiable. Nutanix’s strategy ensures that enterprises can maintain security and performance standards while optimizing costs, according to Caswell.

“For customers who are trying to get started, they’re trying to figure out how do I get started and don’t get locked into pieces that I might want to change over time,” he explained. “That’s where we offer choices of [large language models], choices of [graphics processing units], choices for running on-prem or in the public cloud, and then you can run where you’re most confident around security performance and even costs.”

Driving value through collaboration: Partnerships as a force multiplier

Partnerships are a cornerstone of Nutanix’s enterprise AI strategy, enabling the company to deliver pre-trained models, infrastructure enhancements and Kubernetes management. Collaborations with Nvidia, Hugging Face and D2iQ Inc. extend the reach and impact of Nutanix’s offerings, accelerating time to value for enterprises, according to Justin Boitano, vice president of enterprise AI at Nvidia.

“What we’re seeing is the fundamental shifts across three vectors happening simultaneously,” Boitano told theCUBE during the Nutanix AI-ready Platform event. “First, the world needs to move away from traditional computing into full-stack accelerated computing. The second thing we see is [that] as people go to full stack accelerated computing, it requires reinvention of the entire infrastructure. All of your software across your state has to change, and all of this has to be done to power this new world of generative AI. Finally, the third vector that we see is agentic AI.”

By partnering with Hugging Face, Nutanix enables access to a vast library of pre-trained models, simplifying enterprises’ development processes. These integrations help companies leverage cutting-edge tools for tasks such as natural language processing and sentiment analysis, according to Thomas Cornely, senior vice president of product management at Nutanix.

“This focus on simplicity, focus on user experience, focus on data and controlling your data, focus on giving you a platform to simply run your applications, that to us were just a natural extension,” Cornely told theCUBE during the company’s AI-Ready Platform event. “If you’re going to … need a platform that allows customers to go and run all their applications and manage that data anywhere, well, you have to do it for their AI-based applications and data … there’s a big need for enterprise customers to get the tooling to stand up those infrastructure platforms to support those new applications.”

Simplifying AI infrastructure for enterprise scale

Operational simplicity lies at the heart of Nutanix’s AI infrastructure strategy, according to Dhaliwal. From predictable pricing models to intuitive deployment tools, Nutanix minimizes the complexity of AI workload management. The company’s internal AI tools, such as Support GPT, Sales GPT and Engineering GPT, drive efficiency across its business functions, according to Mandy Dhaliwal, chief marketing officer of Nutanix.

“AI is a C-suite and boardroom priority,” she told theCUBE during the company’s AI-Ready Platform event. “Fifty-nine percent of companies that we talked to believe it is going to change the way they operate fundamentally.”

By leveraging its own AI tools, Nutanix showcases how automation can enhance speed, accuracy and operational efficiency. This “eat your own cooking” approach demonstrates how AI can transform workflows across customer-facing roles, revenue operations and engineering, according to Dhaliwal.

“We drink our own champagne around here,” she said. “We have a support GPT, we have a sales GPT [and] we have an engineering GPT, and the use cases are all around speed to response and accuracy for customer-facing roles, business process alignment around the revenue operations side, automation of repeatable tasks from an engineering perspective, i.e., creation of unit tests.”

(* Disclosure: Nutanix Inc. sponsored some of the segments of theCUBE referenced above. Neither Nutanix nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.) 

Photo: SiliconANGLE/Bing