Announcing Nutanix Support for AI Performance Acceleration with Intel® AMX
A Flexible New Option for Nutanix Customers to Run Lightweight AI Workloads using Nutanix AOS and AHV
As AI continues to revolutionize IT, Nutanix is excited to announce that the Nutanix Cloud Platform now supports Intel® Advanced Matrix Extensions (Intel® AMX) for Nutanix Cloud Infrastructure AOS and AHV hypervisor. This game-changing solution helps Nutanix customers accelerate AI workloads on their existing Intel® Xeon® infrastructure using CPUs.
Intel AMX empowers Nutanix customers with highly adaptable, AI-ready hardware that can support a range of common AI use cases, enabling faster deployments with improved operational agility to unlock highly efficient AI initiatives. Let’s explore why Intel AMX support marks an important step forward in our commitment to providing flexible, scalable solutions for AI-driven applications.
How Does Intel® AMX Work with Nutanix AOS?
Built into the latest Intel Xeon processor cores (4th Gen Intel Xeon processor or later), Intel AMX is a CPU-level feature that can dramatically improve performance and efficiency for AI workloads, allowing Nutanix customers to support a range of common AI-enriched applications without the need for a GPU or other discrete hardware.
AOS support for Intel AMX offers Nutanix customers an efficient, versatile engine for kickstarting AI projects—while marking a foundational step towards deeper Intel AMX integration with our full-stack AI solutions under development like Nutanix Enterprise AI.
Intel AMX works by optimizing the matrix multiplication that powers AI workloads using two main components:
- Tiles: Eight two-dimensional registers, each 1 kilobyte in size, which store large blocks of data.
Tile Matrix Multiplication (TMUL): An accelerator engine integrated with tiles to perform matrix-multiply computations for AI.
Developers can code AI functionality to take advantage of the Intel AMX instruction set, while still coding non-AI functionality to use the processor instruction set architecture (ISA). You can learn more about how Intel AMX works to maximize performance and efficiency here.
Efficient Adaptability for Lightweight AI Workloads on Nutanix
By accelerating performance directly on the CPU, Intel AMX expands the range of AI workloads that Nutanix customers can support using existing hardware. Compatible with leading data types used for training (BF16) and real-time inference (INT8), Intel AMX shines in running AI workloads like lightweight inference, recommender systems, small language models, and edge analytics, providing the required performance to drive value through responsive new capabilities without the need for specialized hardware that adds to system complexity and cost.
This flexibility can drive improved power efficiency, potentially reducing energy costs, cooling needs, and emissions compared to specialized hardware, all while giving Nutanix customers the adaptability to limit total cost of ownership and preserve scalability for future growth.
Simplified Development for AI Applications
Intel AMX provides Nutanix users with all the capabilities needed to simplify development and was designed in close consultation with open-source projects, including TensorFlow and PyTorch. These popular frameworks are optimized for Intel AMX, facilitating rapid access to the latest optimizations and features from Intel. AMX supports 300+ popular AI large language models (LLMs) with up to 10 billion parameters, in addition to a range of other machine learning (ML) and deep learning (DL) workloads.
Intel® AMX was designed to accelerate targeted AI workloads without requiring organizations to add additional layers of hardware complexity. Now, Nutanix customers have the flexibility to improve AI performance using their existing Intel® Xeon® processors — and that means the freedom to begin scaling up a wide range of AI-driven capabilities, tapping into new value without locking into specialized infrastructure investments.
With Intel® AMX, customers with the latest Intel® Xeon® processors will have access to an unprecedentedly flexible engine for AI workloads. Scaling up powerful yet efficient AI capabilities is a challenge that calls for carefully aligning hardware to specific use cases. Intel AMX will make this task dramatically easier, opening the door to more agile, workflow-aligned deployments at the edge and beyond.
Learn More About Building an Efficient AI-Ready Datacenter
For organizations working to expand AI capabilities while maintaining highly scalable systems, Intel Xeon processors with Intel AMX provide a powerful combination ready to help Nutanix Cloud Platform users enable AI/ML applications across private and hybrid cloud environments.
The Nutanix Cloud Platform solution provides a simple and scalable full-stack solution of infrastructure, data services, and management to extend hybrid AI capabilities wherever your apps and data run — capabilities that thrive on the streamlined, cost-effective, and power-efficient options that Intel Xeon processors with Intel AMX can deliver.
Intel AMX is just one way that the latest generation of Xeon processors are enabling a modern datacenter ready to deliver security, efficiency, and flexible performance for AI-enriched workloads. Please see our blog here to learn more about the benefits of Nutanix Cloud Platform support for the latest Intel Xeon processors.
©2025 Nutanix, Inc. All rights reserved. Nutanix, the Nutanix logo and all Nutanix product and service names mentioned herein are registered trademarks or trademarks of Nutanix, Inc. in the United States and other countries. All other brand names mentioned herein are for identification purposes only and may be the trademarks of their respective holder(s). Certain information contained in this content may link or refer to, or be based on, studies, publications, surveys, and other data obtained from third-party sources and our own internal estimates and research. While we believe these third-party studies, publications, surveys, and other data are reliable as of the date of publication, they have not independently verified unless specifically stated, and we make no representation as to the adequacy, fairness, accuracy, or completeness of any information obtained from a third-party. Our decision to publish, link to or reference third-party data should not be considered an endorsement of any such content.