HPE Cray XD670 Accelerated Server (H100 NVL)

The HPE Cray XD670 is a purpose-built accelerated server engineered for large-scale AI model training and high-performance computing workloads. It accommodates 8x NVIDIA H100 NVL GPUs with full liquid cooling readiness, delivering exceptional compute density for transformer-based neural network training, inference scaling, and scientific simulations. The dual-socket architecture supports 4th Gen Intel Xeon Scalable processors, enabling balanced CPU-GPU resource allocation for complex AI pipelines.

Designed for enterprise data centers pursuing production-grade generative AI infrastructure, the XD670 integrates HPE's systems management and reliability engineering with NVIDIA's latest accelerator technology. The platform supports high-bandwidth GPU interconnect fabrics and advanced thermal management for sustained performance under intensive computational loads. This configuration addresses the infrastructure demands of organizations training and deploying large language models, multimodal AI systems, and data-parallel distributed training frameworks.

Ideal for

• Large language model (LLM) pretraining and fine-tuning at scale
• Generative AI model development and transformer architecture experimentation
• High-performance computing simulations in climate modeling, materials science, and physics
• Enterprise machine learning infrastructure for computer vision and NLP applications
• AI inference serving with real-time throughput requirements
• Distributed deep learning research and academic AI computing centers

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.