GIGABYTE G383-R80 - Dual 5th Gen Intel Xeon Scalable processors

The GIGABYTE G383-R80 (part number G383-R80-AF2) is a 3U server for AI training and HPC, supporting dual Intel Xeon Scalable CPUs, 8TB DDR5, and 8x NVIDIA H100 GPUs. Showcased at COMPUTEX 2025, it’s orderable now for large-scale AI deployments.

get a quote
No items found.

Technical Specifications

Form Factor

3U rack-mount server

CPU

Dual 5th Gen Intel Xeon Scalable processors (up to 144 cores each)

Memory

Up to 8TB DDR5-5600 (32x DIMM slots)

Expansion PCIe

Up to 8x PCIe 5.0 x16 slots (for NVIDIA H100, A100, or Intel Gaudi 3 accelerators)

HD

Up to 24x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSDs

Network Interfaces  

  • Supports NVIDIA ConnectX-8 SuperNICs (400GbE) or NVIDIA Quantum-2 InfiniBand  
  • 2x 10GbE LAN ports (management)

Performance Features  

  • Optimized for AI training, inferencing, and HPC workloads  
  • Supports NVIDIA HGX H100 8-GPU platform with NVLink for large-scale AI models  
  • Features GIGABYTE’s liquid-cooling for energy-efficient AI deployments  
  • Certified for NVIDIA AI Enterprise and NVIDIA NIM microservices

Management  

  • GIGABYTE Management Console (GMC) for unified resource orchestration  
  • Integrates with NVIDIA AI Workbench for AI development

Power Supply  

  • Redundant, hot-swappable 3000W power supplies  
  • Liquid-cooling capable for high-density AI data centers

Target Market

Hyperscalers, enterprises, and research institutions for AI and HPC

Ready?

Our team of experts are available and ready to assist you with any inquiries and offer customized 3-5 year financing options that align with your budget and goals.

Contact us