Unlock real-time ray tracing, AI supercomputing, and hyperrealistic graphics—powered by the NVIDIA ecosystem.
GPU Solutions for the AI Era
From data centers to workstations, NVIDIA accelerates the most demanding workloads.
Lenovo ThinkSystem SR630 V3
The Lenovo ThinkSystem SR630 V3 (part number 7D73CTO1WW) is a 1U server for AI inference and edge AI, supporting dual Intel Xeon CPUs, 4TB DDR5, and 3x NVIDIA or Intel accelerators.
Dual 5th Gen Intel Xeon Scalable processors (up to 144 cores each)
Up to 4TB DDR5-5600 (16x DIMM slots)
Up to 10x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSDs
Up to 3x PCIe 5.0 x16 slots (for NVIDIA A100, L40S, or Intel Gaudi 3 accelerators)
Lenovo ThinkSystem SR650 V3
The Lenovo ThinkSystem SR650 V3 is a 2U server for AI training and HPC, supporting dual Intel Xeon CPUs, 8TB DDR5, and 4x NVIDIA H100 GPUs. Highlighted at NVIDIA GTC 2025, it’s orderable now for scalable AI infrastructure.
Dual 5th Gen Intel Xeon Scalable processors (up to 144 cores each)
Up to 8TB DDR5-5600 (32x DIMM slots)
Up to 20x 3.5” or 40x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSDs
Up to 8x PCIe 5.0 x16 slots (for NVIDIA H100, A100, or L40S GPUs)
GIGABYTE R284-A91
The GIGABYTE R284-A91 (part number R284-A91-AAL3) is a 2U server for memory-intensive AI workloads, supporting dual Intel Xeon CPUs, 4TB DDR5, CXL memory, and 4x NVIDIA or Intel accelerators.
Dual Intel Xeon 6900 Series processors (up to 144 cores each)
Up to 4TB DDR5-5600 (24x DIMM slots), plus 16x Micron CXL memory modules
Up to 12x E3.S NVMe hot-swap bays, plus 2x M.2 SSDs
Up to 4x PCIe 5.0 x16 slots (for NVIDIA A100 or Intel Gaudi 3 accelerators), plus CXL expansion
GIGABYTE G493-SB0
The GIGABYTE G293-Z40 (part number G293-Z40-AF2) is a 2U server for AI inference and edge AI, supporting a single AMD EPYC CPU, 3TB DDR5, and 4x NVIDIA GPUs. Orderable now, it’s ideal for compact AI setups.
Dual 5th Gen Intel Xeon Scalable processors (up to 144 cores each)
Up to 4TB DDR5-5600 (16x DIMM slots)
Up to 12x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSDs
Up to 8x PCIe 5.0 x16 slots (for NVIDIA L40S, A100, or AMD Instinct MI300X GPUs)
GIGABYTE G894-SD1
The GIGABYTE G894-SD1 (part number G894-SD1-AAX5) is a 4U server for AI training and inference, supporting dual Intel Xeon CPUs, 8TB DDR5, and 12x NVIDIA B200 or Intel Gaudi 3 accelerators.
Dual Intel Xeon 6700 or 6500 Series processors (up to 128 cores each)
Up to 8TB DDR5-5600 (32x DIMM slots)
Up to 24x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSD.
Up to 12x PCIe 5.0 x16 slots (for NVIDIA HGX B200, H100, or Intel Gaudi 3 accelerators)
GIGABYTE S183-SH0 - Single AMD EPYC 9004 Series processor (up to 96 cores)
The GIGABYTE S183-SH0 (part number S183-SH0-AF1) is a 1U edge server for AI inference, supporting a single Intel Xeon CPU, 2TB DDR5, and NVIDIA A2/T4 GPUs.
Single 5th Gen Intel Xeon Scalable processor (up to 144 cores)
Up to 2TB DDR5-5200 (16x DIMM slots)
Up to 4x 2.5” NVMe/SATA hot-swap bays, plus 2x M.2 SSDs
2x PCIe 5.0 x16 slots (for NVIDIA A2 or T4 GPUs)
GIGABYTE G293-Z40
The GIGABYTE G293-Z40 (part number G293-Z40-AF2) is a 2U server for AI inference and edge AI, supporting a single AMD EPYC CPU, 3TB DDR5, and 4x NVIDIA GPUs. Orderable now, it’s ideal for compact AI setups.
Single AMD EPYC 9004 Series processor (up to 96 cores)
Up to 3TB DDR5-4800 (12x DIMM slots)
Up to 12x 2.5” NVMe/SATA hot-swap bays, plus 2x M.2 SSDs
Up to 4x PCIe 5.0 x16 slots (for NVIDIA A100 or L40S GPUs)
GIGABYTE G593-ZD2 Up to 6TB DDR5-4800
The GIGABYTE G593-ZD2 (part number G593-ZD2-AF2) is a 5U server for AI training and inference, supporting dual AMD EPYC CPUs, 6TB DDR5, and 10x NVIDIA or AMD GPUs. Orderable now, it’s ideal for AI factories with liquid-cooling.
Dual AMD EPYC 9004 Series processors (up to 96 cores each)
Up to 6TB DDR5-4800 (24x DIMM slots)
Up to 12x 3.5” or 24x 2.5” NVMe/SATA/SAS hot-swap bays
Up to 10x PCIe 5.0 x16 slots (for NVIDIA H100, A100, or AMD Instinct MI350 GPUs)
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.
Dual 5th Gen Intel Xeon Scalable processors (up to 144 cores each)
Up to 8TB DDR5-5600 (32x DIMM slots)
Up to 24x 2.5” NVMe/SATA/SAS hot-swap bays, plus 2x M.2 SSDs
Liquid cooled NVIDIA HGX™ H100 with 8 x SXM5 GPUs 8 x PCIe x16 (Gen5 x16) low-profile slots
Lenovo ThinkSystem SR675 V3
2U AI-optimized server supporting dual AMD EPYC 9004 CPUs, up to 6TB DDR5, and 8x NVIDIA H100 GPUs.
Dual AMD EPYC 9004 Series processors (up to 96 cores each)
Up to 6TB DDR5-4800 (24x DIMM slots)
Up to 24x 2.5” NVMe/SATA/SAS hot-swap bays or 12x 3.5” bays
Up to 8x PCIe 4.0/5.0 x16 slots (for NVIDIA H100, A100, or L40S GPUs)
Lenovo ThinkSystem SR685a V3
The SR780a V3 combines 8x latest NVIDIA GPUs with 5th Gen Xeon CPUs in a liquid-cooled 5U chassis for generative AI workloads
Supports up to 2 5th Gen Intel Xeon Scalable processors
Up to 96 GB per DIMM
8 x 2.5-inch hot-swap drive bays supporting PCIe 5.0 NVMe drives
8 x PCIe 5.0 x16 FHHL
Lenovo ThinkSystem SR780a V3
The SR780a V3 combines 8x latest NVIDIA GPUs with 5th Gen Xeon CPUs in a liquid-cooled 5U chassis for generative AI workloads
Supports up to 2 5th Gen Intel Xeon Scalable processors
Up to 96 GB per DIMM
8 x 2.5-inch hot-swap drive bays supporting PCIe 5.0 NVMe drives
8 x PCIe 5.0 x16 FHHL
Dell PowerEdge XE8545
Memory Module Slots: 32 slots (288-pin), 16 slots per processor.
Supports up to 2 AMD EPYC 7003 Series processors (3rd Generation, Milan), with up to 64 cores per CPU
8/16/32 DIMMs (2TB max with XE9640 GPU configuration)
Supports up to 10 x 2.5-inch hot-swappable drives
Supports up to 4 PCIe Gen 4 expansion slots
Dell PowerEdge XE9640
High-density 6U server supporting 8x GPUs for AI acceleration.
p to 2 Intel Xeon Scalable processors (4th or 5th Generation
8/16/32 DIMMs (2TB max with XE9640 GPU configuration)
Up to 4 x 2.5-inch NVMe SSD drives
Up to 4 PCIe Gen 5 slots (x16), designed for high-performance expansion
HPE ProLiant DL380 Gen11
The HPE Apollo 6500 Gen11 Server features dual 4th/5th Gen Intel Xeon or AMD EPYC processors, supporting 8x high-performance GPUs for AI and HPC workloads.
DIMM Speed: Up to 5600 MT/s (DDR5)
DIMM Speed: Up to 5600 MT/s (DDR5)
24x NVMe U.2/EDSFF (368TB max) + dual M.2 boot drives.
PCIe 5.0 x16 slots + OCP 3.0 (200Gbps) for GPUs/NVMe/network expansion.
HPE Apollo 6500 Gen11
The HPE ProLiant DL380 Gen11 features dual Intel Xeon or AMD EPYC processors, supporting 4x GPUs for AI and virtualization.
Intel 4th/5th Gen Xeon Scalable (up to 64 cores) or AMD EPYC 9004 (up to 96 cores).
Memory Type: RDIMM/3DS RDIMM
32x NVMe U.2/EDSFF (max 491TB) + dual M.2 boot drives.
PCIe 5.0 x16 slots + OCP 3.0 (200Gbps) for GPUs/NVMe expansion.
PowerEdge XE8640 Rack Server
The PowerEdge XE8640 Rack Server features four NVIDIA® H100 GPUs, 4 TB max RDIMM memory, up to 122.88 TB storage with NVMe SSDs, and comprehensive security measures.
up to 56 cores per processor
Up to 4800 MT/s, RDIMM 4 TB max
Up to 8 x 2.5-inch NVMe SSD drives max 122.88 TB
2 CPU configuration: Up to 4 PCIe slots (4 x16 Gen5)
PowerEdge XE9640 Rack Server
The Dell PowerEdge XE9640 Rack Server features dual 4th Generation Intel® Xeon® Scalable processors, offering up to 56 cores per processor.
up to 56 cores per processor
RDIMM 1 TB max, Up to 4800 MT/s
Up to 4 x 2.5-inch NVMe SSD drives max 61.44 TB
2 CPU configuration: Up to 4 PCIe slots (4 x16 Gen5)
PowerEdge XE9680 Rack Server
The Dell PowerEdge XE9680 rack server is a high-performance, robust server designed to meet data-intensive workloads and complex computing tasks.
Two 4th Generation Intel® Xeon® Scalable processor with up to 56 cores per processor
32 DDR5 DIMM slots, up to 4800MT/s
Up to 8 x 2.5-inch NVMe SSD drives max 122.88 TB
Up to 10 x16 Gen5 (x16 PCIe) full-height, half-length
SuperServer SYS-421GU-TNXR | Universal 4U Dual Processor
Ideal for High-Performance Computing, AI/Deep Learning, LLM NLP with 8TB DDR5 memory, 8 PCIe Gen 5.0 slots, flexible networking, 2 M.2 NVMe/SATA, and 6 hot-swap drive bays.
Dual 5th Gen Intel® Xeon® / 4th Gen Intel® Xeon® Scalable processors
Slot Count: 32 DIMM slots Max Memory: Up to 8TB 5600MT/s ECC DDR5
Hot-swap : 6x 2.5" hot-swap NVMe/SATA drive bays (6x 2.5" NVMe hybrid)
1 PCIe 5.0 x16 LP slot(s) 7 PCIe 5.0 X16 slot(s)
A+ Server AS-4125GS-TNRT
AI/Deep Learning optimized, supports 8 GPUs, dual AMD EPYC™ 9004 up to 360W, 6TB DDR5 memory, 4 hot-swap NVMe bays, 1 M.2 slot, AIOM/OCP 3.0 support. Ideal for high-performance computing.
Dual AMD EPYC™ 9004 Series Processors up to 360W TDP
Slot Count: 24 DIMM slots Max Memory: Up to 6TB 4800MT/s ECC DDR5 RDIMM/LRDIMM
24x 2.5" hot-swap NVMe/SATA/SAS drive bays (4x 2.5" NVMe dedicated)
9 PCIe 5.0 x16 FHFL slot(s)
SuperServer SYS-521GE-TNRT 5U Dual Processor
Meet the GPU SuperServer SYS-521GE-TNRT: A high-performance server with 5th/4th Gen Intel® Xeon® support, 8TB DDR5 memory, 13 PCIe Gen 5.0 slots, AIOM/OCP 3.0, and 8 hot-swap SATA bays. Ideal for data-intensive tasks.
Up to 64C/128T; Up to 320MB Cache per CPU
Max Memory (2DPC): Up to 8TB 4400MT/s ECC DDR5 RDIMM
8 x 2.5" Gen5 NVMe/SATA/SAS hot-swappable bays
Up to 10 double-width or 10 single-width GPU(s)
5688A7 Extreme AI Server Delivering 16 PFLOPS Performance
5688A7 is an advanced AI system supporting NVIDIA HGX H100/A100 8-GPUs to deliver industry-leading 16PFlops of AI performance.
2x AMD EPYC™ 9004 Series processors
32x DDR5 DIMMs, up to 4800MT/s
24x 2.5” SSD, up to 16x NVMe U.2 2x Onboard NVMe/SATA M.2
5468M7 8 GPU AI Server Supporting Front/Back I/O & Liquid Cooling
5468M7 is a powerful 4U AI server with the latest PCIe Gen 5.0 architecture, supporting 8 FHFL PCIe GPU cards and 4th Gen Intel® Xeon® Scalable processors.
2x 4th Gen Intel® Xeon® Scalable Processors, up to 350W
32x DDR5 4800MT/s RDIMMs
24x 2.5” SSD, up to 16x NVMe U.2 2x NVMe/SATA M.2
8x FHFL double-width PCIe 5.0 GPU cards Supports additional 2x FL double-width PCIe GPU, 1x FL single-width PCIe
5468A7 8 GPU AI Server with PCIe 5.0
5468A7 is a powerful 4U AI server supporting AMD EPYC™ 9004 processors and the latest PCIe Gen 5.0 architecture. Its 10-card pass-through design significantly reduces CPU and GPU communication latency.
2x AMD EPYC™ 9004 processors, up to 400W
24x DDR5 4800MT/s RDIMMs
24x 2.5” SSD, up to 16x NVMe U.2 2x NVMe/SATA M.2
11x PCIe 5.0 x16 slots
GPU Solutions for the AI Era
From data centers to workstations, NVIDIA accelerates the most demanding workloads.
AI in Healthcare
Next-Gen Gaming
Expert recommendations for Your Workflow
Get a Quote
