HP Z8 Fury G5 Workstation, Intel Xeon w9-3495X

The HP Z8 Fury G5 Workstation (5F133UP) is HP's flagship tower workstation built around the Intel Xeon w9-3495X processor, a 60-core, 120-thread Sapphire Rapids-AP chip delivering up to 4.9 GHz maximum turbo frequency. The system is engineered around HP's Z8 Fury platform, which provides a PCIe 5.0 backbone, support for up to 8 TB of DDR5 ECC registered memory across 16 DIMM slots, and a tool-less chassis architecture designed for rapid field reconfiguration. The platform supports up to four double-width professional or compute GPUs simultaneously, making it one of the most GPU-dense single-socket tower workstations commercially available.

The Z8 Fury G5 is purpose-built to handle memory- and compute-intensive workloads that exceed the capacity of conventional workstations. The Xeon w9-3495X's 112.5 MB of total Intel Smart Cache and its high-bandwidth memory subsystem sustain throughput for large model inference, multi-stream simulation, and full-resolution visual effects pipelines without the latency penalties associated with memory bandwidth bottlenecks. The chassis accommodates high-TDP GPU configurations through a robust internal power delivery architecture and active thermal management, supporting professional graphics and GPU compute cards from NVIDIA RTX and AMD Radeon Pro lineups.

Targeted at rendering studios, AI research labs, and computational science environments, the Z8 Fury G5 runs HP Performance Advisor and is certified for leading independent software vendor applications including Autodesk, Dassault Systèmes, and Siemens PLM suites. HP's Z Turbo Drive storage options using PCIe 5.0 NVMe slots provide the sequential read bandwidth needed to feed multi-GPU pipelines without storage-side starvation. The system ships with Windows 11 Pro for Workstations and is also available in Linux-ready configurations, fitting naturally into enterprise IT environments with existing software asset management and imaging workflows.

Ideal for

• Multi-GPU deep learning model prototyping and iterative training using frameworks such as PyTorch and TensorFlow on up to four high-memory compute GPUs
• High-fidelity 3D rendering and real-time ray tracing for visual effects production, architectural visualization, and product design review
• Large-scale finite element analysis and computational fluid dynamics simulations requiring sustained memory bandwidth and high core counts
• Medical imaging processing and AI-assisted diagnostics development where ECC memory integrity and ISV certification are mandatory requirements
• Seismic data processing and reservoir simulation in upstream energy sectors demanding parallel floating-point throughput across many CPU cores
• Multi-disciplinary CAD and CAE workflows combining mechanical simulation, electronics design, and generative design in a single certified workstation platform

Technical specifications

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