Make the amazing possible at any scale. Intel® Gaudi® 3 AI accelerators support state-of-the-art generative AI and LLMs in the data center and pair with Intel® Xeon® processors—the host CPU of choice for leading AI systems —to deliver enterprise performance and reliability.
Intel® Gaudi® 3 Al accelerators and Intel® Gaudi® 3 software are designed to bring a new level of productivity advantages and choice to data center generative Al.
Save Time and Power
TRAIN MODELS IN LESS TIME
1.5x
faster than NV H100 on average
OUTPUT RESULTS FASTER
1.5x
faster inference than NV H100 on average
USE LESS POWER
1.4x
higher inference power efficiency than NV H100 on average
Learn more what Intel Gaudi processors can deliver for you.
Performance and Efficiency at Every Scale
With the surging demand for the advantages deep learning and generative Al can bring, there's never been a greater need for improved computing performance, efficiency, usability, and choice. Intel® Gaudi® 3 Al accelerators and software are designed to bring a new level of computing advantages and choice to data center training and inference-whether in the cloud or on-premises. We aim to make the benefits of Al deep learning more accessible to more enterprises and organizations, removing barriers to bring deep learning advantages to many.
High-Efficiency, Deep Learning-Optimized Processors and Software.
To bring Al to many, the Intel Gaudi platform was conceived and architected to address the training and inference demands of large-scale era Al, providing enterprises and organizations with high-performance, high-efficiency deep learning computing.
Efficient Performance
Architected expressly for DL compute
Heterogeneous computing architecture
AI-optimized matrix multiplication engine
Custom Tensor Processor cores
Large on-board memories
Networking integrated on chip
Massive Flexibility and Scale
On-chip integration of industry-standard RoCE
Massive capacity with integration of 24x 200GbE ports
All-to-all configuration within the server
Flexible scale-out to support numerous configurations
Industry-standard networking lowers cost
Avoids vendor lock-in
Ease of Model Migration & Build
Software optimized for Deep Learning training & inference
Integrates popular frameworks: TensorFlow and PyTorch
Provides custom graph compiler
Supports custom kernel development
Enables ecosystem of software partners
Habana GitHub & Community Forum
Designed for the Real-World Demands of AI
Intel® Gaudi® 3 AI accelerators empower you to use open, community-based software and industry-standard Ethernet networking to scale systems more flexibly.
Built for training and inference
64
Tensor Processor Cores (5th gen)
8
Matrix math engines
Increased memory for LLM efficiency and cost-effectiveness
128 GB
HBM capacity, 3.7 TB/s B/W
96 MB
SRAM, 12.8 TB/s SRAM B/W
Massive, flexible on-chip networking
Open standard vs. proprietary InfiniBand
24x 200GbE
Industry-standard RoCE Ethernet ports
PCle 5
x 16
The best processors to meet your diverse performance and efficiency requirements
The Intel® Xeon® 6 processor family introduces a robust computing platform that excels at both performance and efficiency, crucial for meeting the evolving demands of modern data centers. From powering compute-intensive AI to enabling scalable cloud-native microservices, the processor family provides versatility for diverse operational requirements.
Intel® Xeon® 6 processor compared to 5th Gen Intel® Xeon® Scalable processor
General compute
Up to 2x higher performance for integer and floating point throughput
Artificial intelligence
Up to 2x higher GenAI performance with BF data types
HPC
Up to 2.3x higher HPC performance based on the industry-standard HPCG benchmark
Microservices
Up to 1.5x better performance/watt for server-side Java throughput
Analytics
Up to 1.6x better performance/watt for MySQL OLTP
Media
Up to 1.5x higher AVC performance/watt
Built for Data Center, Cloud, Networking and Edge Deployments
Workload and Performance Tested Across Industry Use Cases
Built on advanced leading-edge process technologies
Enables Confidential Computing and Enhanced Security
Intel® Xeon® 6 Prioritized Workloads and Use Cases
AI
HPC
Infrastructure & Storage
Data Services
Networking and Media
Web, App & Microservices
Computer vision
Earth System
Storage (incl. HCI)
Big Data Analytics
Cloud Gaming
Enterprise Applications (incl. ERP, CRM, SCM)
Image recognition and classification
Financial Services
Virtual desktop infrastructure (VDI)
Business Intelligence (OLAP)
Immersive media
Consumer Applications
Natural Language Processing (incl. Gen AI)
Life & Material Sciences Manufacturing
Security
Datalake
Media Analytics
Web Hosting / Content management system
Recommendation Systems
Data Preparation
Content delivery network (CDN)
Robotic process automation (RPA)
Relational Database
Live Media Processing
Nonrelational Databases
Networking (incl. VRAN, 5G, UPF)
OLTP
Hover the mouse to learn more about each content
AI
AI
Integrate more AI solutions into business processes or new applications with enhanced compute capabilities and built-in AI accelerator engines.
HPC
HPC
Accelerated computing performance and high memory bandwidth help speed up time to insights for models and large-scale calculations.
Storage
Storage
Speed data throughput with fewer cores to improve performance for hyper-converged infrastructure and storage solutions.
Analytics
Analytics
Accelerate your data analytics pipeline to handle more transactions and improve database, CRM, and business intelligence application performance.
Networking
Networking
Accelerate the delivery of services and support more users, transactions, and data with improved network efficiency, packet processing, transcoding, and encryption.
Cloud Computing
Cloud Computing
Improving microservices throughput for exceptional quality of service, overhead, and observability.
Edge
Edge
Deploy edge applications quickly and get critical insights to drive business value closer to where data is generated.
Features and Benefits of Intel Xeon processors
Servers powered by Xeon processors are often found performing workload-heavy computation for cloud computing data centers, analytical and trending applications, radar systems, industrial manufacturing, intelligence-gathering programs and much more.
Enhanced performance
Performance for a broad number of workloads.
Workload optimized
Optimized offerings for networking, cloud, and single socket use cases.
Enhanced security
Capabilities that help address current and future privacy and security concerns.
Optimized processor
Specifications and power requirements support space-and power constrained environments.
Intel Xeon Scalable processors
Intel Xeon Scalable processors feature built-in accelerators and advanced security technologies for the most in-demand workload requirements — all while offering the greatest cloud choice and application portability
Intel® Xeon® Platinum Processors
Advanced 2, 4 & 8 socket performance, designed for the most demanding workloads & services from the edge to cloud.
Intel® Xeon® Gold Processors
Up to 4 socket scalable performance, advanced reliability, and advanced security solutions.
Intel® Xeon® Silver Processors
Performance and power efficiency for entry compute, network and storage.
Intel® Xeon® Bronze Processors
Reliability and serviceability for small business and storage server solutions.
