HPE Cray XD670

Accelerate AI performance for Large Language Model training, Natural Language Processing, and multimodal training.

View the explainer video
HPE Cray XD670.
Boost AI performance for GPU-intensive workloads.

Boost AI performance for GPU-intensive workloads

Develop and train larger AI models faster and accelerate time-to-results with unprecedented efficiency, precision, and detailed insights.

View the demo

Scale to accommodate growing AI workloads

Leverage breakthrough innovation with fully integrated solutions from the hardware to the software layer to take on workloads of any size and scope.

Accelerate AI initiatives

Deploy AI at scale with a solution optimized for workloads in Natural Language Processing (NLP), Large Language Model (LLM) training, and multimodal training.

Meet evolving business demands with optimum flexibility

Improve power efficiency with a direct liquid cooling option and create a flexible environment with a broad range of supported technologies including accelerators, storage and networking.

View the solution brief

Our customers

  • Crusoe Energy

    Together with HPE, Crusoe is building modular data centers using a GPU-accelerated supercomputing platform that can scale as needed in a standard rack format.

    View the case study

Boost AI performance with the #1 server for natural language processing

HPE Cray XD670 is #1 in NLP and a top performer across all the MLPerf Inference v4.0 benchmark models where it participated, including GenAI, Computer Vision, and LLMs.

View the article

5 reasons to choose HPE Cray XD670

Organizations can greatly benefit from the scale and power of a system designed and optimized for AI workloads that are heavily parallelized, requiring GPU acceleration for optimum performance.

View the brochure

Empower high performance image processing

HPE Cray XD670 server enhances performance and reduces processing time for CryoSPARC™ software, used to perform data analysis and 3D reconstruction of cryogenic electron microscopy (cryo-EM) scans.

Learn more
Portrait of influencer having a break between meetings.

Take the next steps

Ready to get started? Explore purchasing options or engage with HPE experts to determine the best solution for your business needs.

Chat now Chat now

More ways to explore

Unlock AI

Simplify AI complexity, accelerate productivity, and get pilots to production faster.

Learn more

HPE Supercomputing

Empower world-changing innovation and discovery in the Exascale era and beyond, with faster time-to-results and accelerated AI.

Learn more

HPE Machine Learning Development Environment Software

Easily implement and train Machine Learning models by removing complexities, optimizing cost, and accelerating innovation.

Learn more
HPE GreenLake cloud platform software screenshot.

Learn more about the HPE GreenLake cloud experience

All HPE GreenLake cloud services are accessed through a unified control plane that delivers a consistent, open and extensible cloud operating experience for all your services and users, wherever the workloads and data are located.

Explore the HPE GreenLake cloud experience

Technical specifications

View the QuickSpecs

  • Chassis

    • 5U chassis system
    • Single 2x CPU node

  • GPU acceleration – Option A

    • 8x NVIDIA H200 SXM 700W TDP GPUs with 141 GB HBM each
    • 2x 5th Gen Intel Xeon Scalable processors CPUs, up to 400W TDP
    • 32x DIMMs DDR5, 8-channel memory per socket, up to 5600 MHz

  • GPU acceleration – Option B

    • 8x NVIDIA H100 SXM 700W TDP GPUs with 80 GB HBM each
    • 2x 4th Gen Intel Xeon Scalable processors CPUs, up to 300W TDP
    • 32x DIMMs DDR5, 8-channel memory per socket, up to 4800 MHz

  • Power and cooling

    • Air cooled and also available with a Direct Liquid Cooling option for improved power efficiency