403
Sorry!!
Error! We're sorry, but the page you were looking for doesn't exist.
Nutanix Gives Enterprises The Freedom To Run Production Agentic AI Their Way
(MENAFN- Mid-East Info) NEWS SUMMARY:
-
Nutanix Cloud Platform NCP is being enhanced and expanded for production agentic AI with its dual-native architecture, including the introduction of Nutanix Enterprise AI NAI 2.8 and Nutanix Kubernetes Platform (NKP) 2.19.
-
The new capabilities give customers a flexible cloud operating model designed to consistently manage and govern AI across environments, supported by a dual-native architecture for virtual machines and containers.
-
NAI 2.8 is available now. It provides centralized control for AI inference and agentic AI, including Nutanix Agent Gateway, now with a generally available Model Context Protocol (MCP) gateway for governing how agents connect with apps and data via MCP. Nutanix Private Inference also provides enhanced capabilities for high-performance fine tuning and inference, along with improved security and governance.
-
NKP 2.19 will be available soon. It is expected to provide streamlined container management for bare metal and virtualized environments, with a built-in AI catalog designed for building and running agentic AI applications.
-
The new capabilities in NAI 2.8 and NKP 2.19, combined with Cloud Native Computing Foundation (CNCF) certification, enhance NCP dual-native and AI capabilities.
-
Nutanix: Verified Services program and Service Provider (SP) Central are now available to help partners drive new AI opportunities.
-
Agent Gateway: This now includes a generally available MCP Gateway which serves as a secure, unified front door for AI agents to access tools and data without custom engineering. To complement this, Nutanix has also released MCP Server for NCP to help customers build agentic AI applications with secure access to the infrastructure managed by Nutanix.
-
Private Inference: New advanced inference and fine tuning capabilities enable scalable, multiGPU inference for LLMs via tensor parallelism, delivering high-throughput serving and low-latency response times for enterprise LLM workloads. In addition, this enables batch inference and speculative decoding. Key features include: 1) Parameter-Efficient Fine-Tuning which supports Low-Rank Adaptation (LoRA) fine-tuning for smaller models (<8B parameters), helping organizations to cost-effectively customize open LLMs on private domain data using single-GPU compute while seamlessly deploying adapters straight to serving pipelines; 2) Scalable multiGPU serving which enables high-throughput multiGPU inference via tensor parallelism, delivering fast, distributed serving across enterprise hybrid cloud environments; and 3) Speculative decoding which accelerates LLM inference token generation by up to 2.5x using lightweight draft models, cutting output latency without sacrificing model accuracy.
Enhanced Security against Rogue AI: With the rise of agentic AI and the risk of models breaking out of sandboxes, security is paramount. NAI provides robust protection against rogue models through our platform and APIs, featuring fine-grained Identity and Access Management (IAM), custom roles and seamless model sharing. This enforces least-privilege security, helping ensure agents operate securely and restricting access to only authorized roles, as well as support for air-gapped NVIDIA NIM deployment.
-
NKP Metal: Built to bring HCI-grade simplicity to bare-metal Kubernetes, with automated OS, firmware, and container deployment, and persistent, enterprise-grade storage natively, eliminating the complexity of patchwork platforms.
NKP Full Stack: While NKP Metal is intended to bring simplicity to bare-metal deployments, NKP on AHV remains the cornerstone for organizations requiring robust, agile virtualized environments. Combined with Nutanix Flow, NKP on AHV is designed to deliver stronger network-level sandboxing for AI agents, helping provide essential isolation to mitigate the risk of rogue attacks and lateral movement.
AI Applications Catalog: Offers a one-click deployment path for curated, validated AI/ML software (Kubeflow, Milvus, Slurm) to help bypass manual integration challenges.
Hardware and Compliance: Planned expansion of ecosystem support with validated GPU integrations, alongside dynamic resource allocation for modern AI workloads.
-
Jasim Abdul Rahman, Group Chief Information Officer, Power International Holding
-
Matt Flug, IDC
-
Jeremy Foster, Cisco GM & SVP, Cisco Compute
-
Edward O'Connor, Chief Technology Officer at Continent 8 Technologies
-
Bill Pearson, VP, Data Center Software, Intel
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept any
responsibility or liability for the accuracy, content, images, videos,
licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright issues
related to this article, kindly contact the provider above.

Comments
No comment