Intel Unveils Xeon 6 and Gaudi 3 AI Chips to Enhance AI and High-Performance Computing (HPC) Capabilities

Intel Launches New Xeon 6 Processors and Gaudi 3 AI Accelerators to Boost AI Performance

Intel has introduced the Xeon 6 processors featuring Performance-cores (P-cores) alongside the Gaudi 3 AI accelerators to strengthen its foothold in the AI landscape. The Xeon 6 processors are designed to double AI vision performance, while the Gaudi 3 accelerators offer 20% enhanced throughput.

Meeting Demand for AI Infrastructure

As AI revolutionizes various industries, companies increasingly require cost-effective infrastructure that supports rapid development and deployment. In response, Intel’s new Xeon 6 and Gaudi 3 solutions are engineered for optimal performance per watt, addressing the demand for powerful AI systems and lowering the total cost of ownership (TCO).

“Demand for AI is transforming data centers, and the industry seeks diverse hardware, software, and developer tools,” stated Justin Hotard, Intel’s Executive Vice President and General Manager of the Data Center and AI Group. “With the launch of Xeon 6 and Gaudi 3, Intel is creating an open ecosystem for enhanced performance, efficiency, and security across workloads.”

Introducing Intel Xeon 6 and Gaudi 3 AI Accelerators

Intel's latest innovations include significant updates to its data center portfolio with the Xeon 6 processors. Designed for compute-intensive workloads, these processors deliver twice the performance of their predecessors, featuring increased core counts, double memory bandwidth, and embedded AI acceleration for optimal efficiency from edge to cloud.

The Gaudi 3 AI accelerator is optimized for large-scale generative AI, equipped with 64 Tensor Processing Cores (TPCs) and eight Matrix Multiplication Engines (MMEs). It supports 128 GB of HBMe2 memory for training and inference and includes 24 200 Gigabit Ethernet ports for scalable networking. Additionally, Gaudi 3 is fully compatible with the PyTorch framework and advanced Hugging Face transformer models. Intel has partnered with IBM to deploy Gaudi 3 AI accelerators as a service on IBM Cloud, aiming to reduce TCO and enhance AI performance.

Enhancing AI Systems with TCO Benefits

Intel's Xeon 6 and Gaudi 3 enhancements are designed with scalable AI deployment in mind, offering flexible options, competitive pricing, and accessible AI technologies. With 73% of GPU-accelerated servers relying on Intel Xeon as the host CPU, Intel's x86 infrastructure and open ecosystem are poised to support enterprises in creating high-value AI systems.

Intel collaborates with leading OEMs, including Dell Technologies, Hewlett Packard Enterprise, and Supermicro, to co-engineer systems that meet specific customer needs for efficient AI deployments. Dell Technologies is currently developing retrieval-augmented generation (RAG)-based solutions using Xeon 6 and Gaudi 3.

Transitioning generative AI solutions from prototypes to production-ready systems involves addressing challenges like real-time monitoring, error handling, logging, security, and scalability. Intel is working closely with OEMs to deliver production-ready RAG solutions built on the Open Platform Enterprise AI (OPEA) framework, integrating microservices for a streamlined approach optimized for Xeon and Gaudi systems.

Expanding Access to Enterprise AI Applications

Intel continues to enhance access to enterprise AI solutions. The Intel Tiber portfolio addresses issues related to access, cost, complexity, security, efficiency, and scalability across AI, cloud, and edge environments. The Intel® Tiber™ Developer Cloud now offers preview systems of the Xeon 6 for technical evaluation and testing.

Select customers will also gain early access to Gaudi 3 for validating AI model deployments, with Gaudi 3 clusters expected to roll out next quarter for large-scale production.

Additionally, new service offerings include SeekrFlow, an end-to-end AI platform from Seekr for developing trusted AI applications. Recent updates feature the latest releases of Intel Gaudi software and Jupyter notebooks, which now support PyTorch 2.4 and Intel oneAPI, enhancing AI acceleration capabilities for Xeon 6 processors.

Most people like

Find AI tools in YBX