TBird-HGX-5 High-Performance GPU Server
The TBird-HGX-5 is an enterprise-grade, ultra-high-density 4U GPU server engineered for artificial intelligence, deep learning, machine learning, and high-performance computing (HPC) workloads. Powered by dual AMD EPYC™ 9005 Series processors and built to house up to 8 dual-slot PCIe Gen5 graphics accelerators, this platform delivers massive compute density and extreme parallel processing capability for demanding enterprise AI tasks.
Featuring advanced multi-channel DDR5 memory support, extensive high-speed PCIe Gen5 interconnects, versatile hot-swappable NVMe storage bays, and high-capacity redundant Titanium power supplies, the TBird-HGX-5 provides a resilient, high-throughput foundation for enterprise AI clusters, large language model training, and advanced data analytics.
Extreme 8-GPU Density: Engineered to support up to 8 dual-slot PCIe Gen5 GPU accelerators for massive acceleration in AI training and deep learning inference.
Dual-Socket Compute Architecture: Powered by dual AMD EPYC™ 9005 Series Processors (Socket SP5) with cTDP up to 500W for ultra-fast multi-threaded processing.
High-Speed DDR5 Memory: Supports high-capacity multi-channel DDR5 RDIMM memory architecture (24 DIMM slots) delivering maximum data throughput.
Advanced Gen5 Interconnects: Features 8 x FHFL PCIe Gen5 x16 slots dedicated for GPUs plus 1 additional FHFL PCIe Gen5 x16 expansion slot.
Versatile Storage Bays: Supports 4 x 2.5" Gen5 NVMe hot-swap drive bays plus onboard M.2 slots for rapid OS boot and ultra-fast data access.
High-Speed Networking: Equipped with dual 10GbE LAN ports (Intel® X710-AT2) and dedicated management LAN for secure remote administration.
Redundant Power Delivery: Equipped with high-capacity 3200W 80 PLUS Titanium redundant power supplies designed to sustain maximum multi-GPU power requirements.
Integrated BMC Remote Management: Integrated ASPEED AST2600 Baseboard Management Controller with dedicated LAN port for out-of-band monitoring and remote administration.
Form Factor
• 4U Rackmount GPU Server
Processor
• Dual AMD EPYC™ 9005 Series Processors
• Socket SP5 (LGA 6096)
• cTDP up to 500W
GPU Expansion
• Supports up to 8 x dual-slot PCIe Gen5 graphics cards / GPU accelerators
Memory
• 24 x DDR5 RDIMM slots
• 12-channel architecture per processor (up to 6400 MT/s)
ExpansionSlots
• 8 x FHFL PCIe Gen5 x16 slots dedicated for full-height dual-slot GPUs
• 1 x additional FHFL PCIe Gen5 x16 expansion slot
Storage
• 4 x 2.5" Hot-Swap Gen5 NVMe drive bays
• 2 x M.2 slots (PCIe Gen3) for system boot
Networking & I/O
• 2 x 10GbE LAN ports (Intel® X710-AT2)
• 1 x Dedicated 1GbE Management LAN port
• Front USB ports and Mini-DP video output via ASPEED AST2600
Power Supply
• Multi-unit 3200W 80 PLUS Titanium redundant power supplies (3+1 / N+N)
The TBird-HGX-5 is engineered for enterprise data centers, AI labs, and high-performance computing environments requiring maximum GPU density for:
• Artificial Intelligence & Deep Learning model training and inference
• Large Language Model (LLM) development and deployment
• High-Performance Computing (HPC) & complex scientific simulations
• Large-scale data analytics, 3D rendering, and GPU cloud infrastructure
• Visual computing and accelerated enterprise workloads
Combining dual AMD EPYC™ 9005 processing, 8 x PCIe Gen5 GPU slot acceleration, high-density DDR5 memory, flexible Gen5 NVMe storage, dual 10GbE networking, and redundant 3200W Titanium power supplies, the TBird-HGX-5 delivers a robust, scalable compute foundation tailored for modern AI and enterprise computing environments.
