Intel® Data Center GPU Max Series

Maximize impact with the Intel® Data Center GPU Max Series, Intel’s highest performing, highest density, general-purpose discrete GPU, which packs over 100 billion transistors into one package and contains up to 128 Xe Cores–Intel’s foundational GPU compute building block.

Product brief: Intel® Data Center GPU Max Series ›

Intel® Data Center GPU Max Series

Key Features

The Intel® Data Center GPU Max Series is designed to take on the most challenging high-performance computing (HPC) and AI workloads. The Intel® Xe Link high-speed, coherent, unified fabric offers flexibility to run any form factor to enable scale up and scale out.

Up to 408MB of L2 Cache

Ensure high capacity and high bandwidth with up to 408 MB of L2 cache (Rambo) based on discrete SRAM technology and 64 MB of L1 cache plus up to 128 GB of high-bandwidth memory.

Built-in Ray Tracing Acceleration

Accelerate scientific visualization and animation with up to 128 ray tracing units incorporated on each Intel® Max Series GPU.

Intel® Xe Matrix Extensions (XMX)

Accelerate AI workloads and enable vector and matrix capabilities in a single device with Intel® Xe Matrix Extensions (XMX) built with deep systolic arrays.

Show more Show less

Maximize Impact

The Intel® Data Center GPU Max Series accelerates science and discovery with breakthrough performance.

Up to

2x

performance gain on HPC and AI workloads over competition due to the Intel® Max Series GPU large L2 cache.1

Up to

12.8x

performance gain over 3rd Gen Intel® Xeon® processors on LAMMPS workloads running on Intel® Max Series CPUs with kernels offloaded to six Intel® Max Series GPUs, optimized by Intel® oneAPI tools.2

Up to

256x

Int8 operations per clock. Speed AI training and inference with up to 256 Int8 operations per clock with the built-in Intel® XMX.

Up to

2x

performance gain on HPC and AI workloads over competition due to the Intel® Max Series GPU large L2 cache.1

Up to

12.8x

performance gain over 3rd Gen Intel® Xeon® processors on LAMMPS workloads running on Intel® Max Series CPUs with kernels offloaded to six Intel® Max Series GPUs, optimized by Intel® oneAPI tools.2

Up to

256x

Int8 operations per clock. Speed AI training and inference with up to 256 Int8 operations per clock with the built-in Intel® XMX.

Intel® Data Center GPU Max Series Demos and Videos

Aurora Blade for Machine Learning Demo

10,000 server blades with the Intel® Max Series processors will power the Aurora supercomputer at Argonne National Laboratory. Preview how a machine learning inference process takes trained AI models to identify structures within a mouse brain to create a 3D visualization.

Aurora Update from Argonne National Laboratory

Rick Stevens, Associate Laboratory Director at Argonne National Laboratory, shares an update on the Aurora Supercomputer status and deployment of Intel® Max Series GPUs.

Materials Science Productivity on LAMMPs

See LAMMPS workloads running on Intel® Max Series CPUs with kernels offloaded to six Intel® Max Series GPUs to achieve performance gains of up to 12.8x over 3rd Gen Intel® Xeon® processors in pursuit of new and improved materials.2

Software and Tools

Discover the open software solutions that are enabling the industry’s most flexible GPUs, including code migration tools and advanced compilers, libraries, profilers, and optimized AI frameworks. For the latest HPC and AI software developer tools, visit Software for Intel® Data Center GPU Max Series.

The Intel HPC Story

Intel’s industry-leading HPC portfolio was designed to fuel the next generation of computing innovation. Optimized to deliver breakthroughs across a range of factors, these technologies help innovators blast through barriers and solve complex problems in entirely new ways.

15+ Intel® Data Center GPU Max Series OEM Designs

Notices and Disclaimers3 4

Compare Products

Product and Performance Information

1

Visit intel.com/performanceindex (Events: Supercomputing 22) for workloads and configurations. Results may vary.

2

LAMMPS (Atomic Fluid, Copper, DPD, Liquid_crystal, Polyethylene, Protein, Stillinger-Weber, Tersoff, Water)

  • Intel® Xeon® 8380: Test by Intel as of 10/11/2022. 1-node, 2x Intel® Xeon® 8380 CPU, HT On, Turbo On, NUMA configuration SNC2, Total Memory 256 GB (16x16GB 3200MT/s, Dual-Rank), BIOS Version SE5C620.86B.01.01.0006.2207150335, ucode revision=0xd000375, Rocky Linux 8.6, Linux version 4.18.0-372.26.1.el8_6.crt1.x86_64, LAMMPS v2021-09-29 cmkl:2022.1.0, icc:2021.6.0, impi:2021.6.0, tbb:2021.6.0; threads/core:; Turbo:on; BuildKnobs:-O3 -ip -xCORE-AVX512 -g -debug inline-debug-info -qopt-zmm-usage=high;
  • 4th Gen Intel® Xeon® Scalable Processor: Test by Intel as of 9/29/2022. 1-node, 2x Intel® Xeon® 8480+, HT On, Turbo On, SNC4, Total Memory 512 GB (16x32GB 4800MT/s, DDR5), BIOS Version SE5C7411.86B.8713.D03.2209091345, ucode revision=0x2b000070, Rocky Linux 8.6, Linux version 4.18.0-372.26.1.el8_6.crt1.x86_64, LAMMPS v2021-09-29 cmkl:2022.1.0, icc:2021.6.0, impi:2021.6.0, tbb:2021.6.0; threads/core:; Turbo:off; BuildKnobs:-O3 -ip -xCORE-AVX512 -g -debug inline-debug-info -qopt-zmm-usage=high;
  • Intel® Xeon® CPU Max Series: Test by Intel as of 9/29/2022. 1-node, 2x Intel® Xeon® Max 9480, HT ON, Turbo ON, NUMA configuration SNC4, Total Memory 128 GB (HBM2e at 3200 MHz), BIOS Version SE5C7411.86B.8424.D03.2208100444, ucode revision=0x2c000020, CentOS Stream 8, Linux version 5.19.0-rc6.0712.intel_next.1.x86_64+server, LAMMPS v2021-09-29 cmkl:2022.1.0, icc:2021.6.0, impi:2021.6.0, tbb:2021.6.0; threads/core:; Turbo:off; BuildKnobs:-O3 -ip -xCORE-AVX512 -g -debug inline-debug-info -qopt-zmm-usage=high;
3Performance varies by use, configuration, and other factors. Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available ​updates. No product or component can be absolutely secure. Learn more at www.Intel.com/PerformanceIndex​.
4

Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy.