Nutanix Enterprise AI Streamlines LLM Deployment and Control in Hybrid Environments

Nutanix, a leader in hybrid multi-cloud computing, is unveiling Nutanix Enterprise AI (NAI), a cloud native offering that simplifies the deployment of large language models (LLMs), on any Kubernetes platform, at the edge, in core data centers, and on public cloud services such as AWS EKS, Azure AKS, and Google GKE. This launch empowers enterprises to innovate in the space of AI quickly, securely, and with hybridity at its core—all while improving ROI, according to Nutanix.

As part of Nutanix’s GPT-in-a-box v2.0—a set of services that provide a turnkey GenAI experience—NAI offers a consistent multi-cloud operating model that streamlines AI operations with resiliency and security. Additionally, NAI’s pricing is transparent and predictable, based on infrastructure resources so that enterprises can effectively anticipate the costs of running the LLMs fueling their GenAI apps.

“Given the span of GenAI/LLMOps, it is important to have consistent operating models to simplify operations. AI and IT admins are often faced with diverse environments across cloud, on-premise data centers and edge, resulting in increased complexity, larger operating costs, etc.,” said Debojyoti Dutta, vice president of engineering (AI) at Nutanix. “With NAI, the IT or the AI admin gets the same interface and user experience independent of where they run the offering. That simplifies people and processes needed to operate AI infrastructure.”

“In this new era of GenAI, LLM models (and AI agents) are first class corporate assets and IT teams need to have a consistent view and control of these enterprise assets,” Dutta continued.

Delivering a streamlined method for securely deploying, scaling, and running LLMs with both NVIDIA NIM optimized inference microservices and open foundation models from Hugging Face, NAI is a user-friendly, centralized GenAI experience for on-prem, the edge, and now for public clouds. Considering the inherent hybridity of GenAI workloads—stretching from the cloud to the private data center to the edge—NAI enables enterprise teams to effectively manage and control their AI deployments.

“NAI gives the user a choice of models (NIM, HuggingFace, Custom) and also the choice of technology/serving engine/stack. This is a unique value proposition for the customer,” said Dutta. “We apply the same consistent operating model to manage the end points, users, access tokens, etc.”

NAI also offers the following advantages:

• Fill AI skill shortage gaps with simple, choice-based, out-of-the-box features that enable IT, data scientists, and developers to easily innovate and accelerate AI development
• Eliminate barriers to building an AI-ready platform with a simple, UI-driven workflow that allows users to deploy and test LLM inference endpoints in as little as minutes, while supporting customer model choice, optimized model performance, and deployment environments
• Mitigate privacy and security risks by enabling enterprises to run models and data on compute resources they control, paired with an intuitive dashboard for troubleshooting, observability, and utilization of resources used for LLMs, and quick and secure role-based access controls
• Deliver enterprise infrastructure to GenAI workloads by delivering the same resiliency, Day 2 operations, and security offered by the Nutanix Cloud Platform to AI initiatives

"Generative AI workloads are inherently hybrid, with training, customization, and inference occurring across public clouds, on-premises systems, and edge locations," said Justin Boitano, vice president of enterprise AI at NVIDIA. "Integrating NVIDIA NIM into Nutanix Enterprise AI provides a consistent multi-cloud model with secure APIs, enabling customers to deploy AI across diverse environments with the high performance and security needed for business-critical applications."

To learn more about NAI, please visit https://www.nutanix.com/.