-
Vendor:NVIDIA
NVIDIA 900-53669-2500-000 | NVLink bridge, 3-slot spacing, 2-way dual-GPU, RTX A6000/A5500/A5000/A4500 compatible
900-53669-2500-000
Vendor:NVIDIANVIDIA 900-53669-2500-000 | NVLink bridge, 3-slot spacing, 2-way dual-GPU, RTX A6000/A5500/A5000/A4500 compatible
Purpose-built NVLink bridge for Ampere-based RTX A-series GPUs with three-slot spacing, enabling fast peer-to-peer links and memory sharing between two identical GPUs. Ideal for data science, rendering, and visualization workloads. -
Vendor:NVIDIA
NVIDIA 900-53669-0000-000 | NVLink bridge, 3-slot spacing, 2-way dual-GPU, RTX A-series professional compatibility
900-53669-0000-000
Vendor:NVIDIANVIDIA 900-53669-0000-000 | NVLink bridge, 3-slot spacing, 2-way dual-GPU, RTX A-series professional compatibility
Dual-GPU NVLink bridge for professional RTX A-series cards installed three slots apart. Provides a high-speed interconnect for memory sharing and faster data exchange in multi-GPU workflows. -
Vendor:NVIDIA
NVIDIA 900-5G192-2551-000 | RTX A2000 Professional Graphics Card, 12GB GDDR6 ECC
900-5G192-2551-000
Vendor:NVIDIANVIDIA 900-5G192-2551-000 | RTX A2000 Professional Graphics Card, 12GB GDDR6 ECC
The NVIDIA RTX A2000 12GB is a professional graphics card built on Ampere architecture, designed for CAD, rendering, and AI-accelerated workflows. It features 3,328 CUDA cores, 26 second-generation RT cores, 104 third-generation Tensor cores, and 12GB of GDDR6 memory with ECC support in a compact dual-slot, low-profile form factor. Ideal for workstations requiring real-time ray tracing, deep learning inference, and professional visualization. -
Vendor:NVIDIA
NVIDIA 900-5G144-2200-000 | RTX PRO 6000 Graphics Card, 96GB GDDR7, 24064 CUDA Cores, 600W
900-5G144-2200-000
Vendor:NVIDIANVIDIA 900-5G144-2200-000 | RTX PRO 6000 Graphics Card, 96GB GDDR7, 24064 CUDA Cores, 600W
The NVIDIA RTX PRO 6000 Blackwell Workstation Edition delivers extreme AI and graphics performance for professionals tackling demanding workflows. Built on the revolutionary Blackwell architecture with 24,064 CUDA cores and 96GB of ECC GDDR7 memory, it accelerates AI development, 3D rendering, real-time simulation, data science, and engineering. This dual-slot, double-flow-through cooled GPU features PCIe 5.0 interface, four DisplayPort 2.1b outputs, and 1792 GB/s memory bandwidth to handle the most complex datasets and multi-app workflows. -
Vendor:Cisco
Cisco UCSAI-880A-B3-SLD | GPU sled for C880A M8, B300 NV variant, NVIDIA accelerator-ready, Granite Rapids platform compatible
UCSAI-880A-B3-SLD
Vendor:CiscoCisco UCSAI-880A-B3-SLD | GPU sled for C880A M8, B300 NV variant, NVIDIA accelerator-ready, Granite Rapids platform compatible
Modular GPU sled engineered for the UCS C880A M8 AI platform. The B300 NV variant accommodates NVIDIA data center accelerator configurations, enabling dense compute offload and high-bandwidth intra-chassis connectivity. Designed for balanced airflow and serviceability within the C880A system. -
Vendor:NVIDIA
NVIDIA 900-5G153-2570-000 | 72GB GDDR7, PCIe 5.0 x16, 4x DisplayPort 2.1, 2-slot workstation GPU
900-5G153-2570-000
Vendor:NVIDIANVIDIA 900-5G153-2570-000 | 72GB GDDR7, PCIe 5.0 x16, 4x DisplayPort 2.1, 2-slot workstation GPU
Advanced professional GPU built on the Blackwell architecture, pairing 72GB of GDDR7 memory with PCIe 5.0 bandwidth and four DisplayPort 2.1 outputs. Engineered for real-time ray tracing, complex visualization, and AI-accelerated content creation in compact two-slot workstations. -
Vendor:Cisco
Cisco UCSX-GPUA100-80-D | GPU | 80GB HBM2e, 300W TDP | PCIe Gen4 x16, passive | MIG-enabled, includes power cable
UCSX-GPUA100-80-D
Vendor:CiscoCisco UCSX-GPUA100-80-D | GPU | 80GB HBM2e, 300W TDP | PCIe Gen4 x16, passive | MIG-enabled, includes power cable
High-performance Ampere-based accelerator delivering 80GB HBM2e for large AI training and inference workloads in UCS X-Series. Passive, server-cooled PCIe Gen4 x16 design with Multi-Instance GPU (MIG) for secure partitioning and consolidation; includes required power cable for installation. -
Vendor:Cisco
Cisco UCSX-GPU-T4MEZZ-D | GPU | 16GB GDDR6, 75W TDP | PCIe Gen3 x16 mezz, passive cooling | vGPU-ready
UCSX-GPU-T4MEZZ-D
Vendor:CiscoCisco UCSX-GPU-T4MEZZ-D | GPU | 16GB GDDR6, 75W TDP | PCIe Gen3 x16 mezz, passive cooling | vGPU-ready
Low-profile, server-cooled mezzanine GPU for UCS X-Series, delivering efficient inference and visualization acceleration in a 75W envelope with 16GB GDDR6. Installs in the front mezz GPU bay to add CUDA-capable compute without requiring external power. Ideal for VDI, AI inference, and analytics at scale. -
Vendor:Cisco
Cisco UCSX-GPU-T4-MEZZ | T4 GPU (Mezz) | 16GB GDDR6 | 75W | UCS X-Series mezzanine | Passive, server-integrated
UCSX-GPU-T4-MEZZ
Vendor:CiscoCisco UCSX-GPU-T4-MEZZ | T4 GPU (Mezz) | 16GB GDDR6 | 75W | UCS X-Series mezzanine | Passive, server-integrated
Mezzanine variant of the 16GB accelerator for UCS X-Series, delivering efficient 75W performance for inference and VDI without occupying standard PCIe slots. Passive design leverages chassis airflow for dense, integrated deployments. -
Vendor:Cisco
Cisco UCSX-GPU-T4-16-D | T4 GPU | 16GB GDDR6 | 75W | PCIe 3.0 x16 | Low-profile, single-slot | Passive
UCSX-GPU-T4-16-D
Vendor:CiscoCisco UCSX-GPU-T4-16-D | T4 GPU | 16GB GDDR6 | 75W | PCIe 3.0 x16 | Low-profile, single-slot | Passive
Low-profile 16GB accelerator optimized for inference, VDI, and media workloads with efficient 75W operation. Designed for passive, high-density deployment in UCS servers without external power connectors. -
Vendor:Cisco
Cisco UCSX-GPU-T4-16 | T4 GPU | 16GB GDDR6 | 75W | PCIe 3.0 x16 | Low-profile, single-slot | Passive
UCSX-GPU-T4-16
Vendor:CiscoCisco UCSX-GPU-T4-16 | T4 GPU | 16GB GDDR6 | 75W | PCIe 3.0 x16 | Low-profile, single-slot | Passive
Versatile low-profile accelerator for AI inference, VDI, and analytics with 16GB GDDR6 and a 75W power envelope. Passive, single-slot design fits dense UCS servers while maintaining excellent performance-per-watt. -
Vendor:Cisco
Cisco UCSX-GPU-MI210 | Instinct MI210 GPU | 64GB HBM2e | 300W | PCIe 4.0 x16 | 2-slot FHFL | Passive
UCSX-GPU-MI210
Vendor:CiscoCisco UCSX-GPU-MI210 | Instinct MI210 GPU | 64GB HBM2e | 300W | PCIe 4.0 x16 | 2-slot FHFL | Passive
High-performance data center accelerator with 64GB HBM2e for HPC and AI workloads. Dual-slot, full-height/full-length passive design for server airflow, delivering PCIe Gen4 bandwidth and robust compute density in UCS environments. -
Vendor:Cisco
Cisco UCSX-GPU-L40S | GPU: L40S, 350W, 48GB GDDR6, PCIe Gen4 x16, FHFL 2-slot
UCSX-GPU-L40S
Vendor:CiscoCisco UCSX-GPU-L40S | GPU: L40S, 350W, 48GB GDDR6, PCIe Gen4 x16, FHFL 2-slot
AI-optimized data center accelerator designed for high-throughput inference, training at scale, and advanced graphics workloads. The L40S combines 48GB ECC GDDR6 with enhanced compute capabilities in a full-height, full-length 2‑slot design for top-tier performance. -
Vendor:Cisco
Cisco UCSX-GPU-L40 | GPU: L40, 300W, 48GB GDDR6, PCIe Gen4 x16, FHFL dual-slot
UCSX-GPU-L40
Vendor:CiscoCisco UCSX-GPU-L40 | GPU: L40, 300W, 48GB GDDR6, PCIe Gen4 x16, FHFL dual-slot
High-performance data center GPU for graphics, rendering, AI training/inference, and accelerated compute. The L40 provides 48GB GDDR6 with ECC and robust Tensor/RT cores in a full-height, full-length design, ideal for demanding workloads and multi-tenant virtualization. This SKU includes an internal power cable for easy integration. -
Vendor:Cisco
Cisco UCSX-GPU-L4-MEZZ | GPU: L4, 70W, 24GB GDDR6, PCIe Gen4 x16, HHHL single-slot
UCSX-GPU-L4-MEZZ
Vendor:CiscoCisco UCSX-GPU-L4-MEZZ | GPU: L4, 70W, 24GB GDDR6, PCIe Gen4 x16, HHHL single-slot
Single-slot HHHL L4 accelerator delivering efficient AI inference, media processing, and virtual desktop density with minimal power draw. PCIe Gen4 x16 bandwidth in a compact footprint makes it ideal for scale-out and edge servers where thermals and space are constrained. -
Vendor:Cisco
Cisco UCSX-GPU-L4 | GPU: L4, 70W, 24GB GDDR6, PCIe Gen4 x16, HHHL single-slot
UCSX-GPU-L4
Vendor:CiscoCisco UCSX-GPU-L4 | GPU: L4, 70W, 24GB GDDR6, PCIe Gen4 x16, HHHL single-slot
Compact, energy‑efficient accelerator optimized for AI inference, video analytics, and VDI density. The L4 delivers modern Tensor/RT core acceleration in a half‑height, half‑length form factor that fits constrained servers while requiring no auxiliary power. Ideal for edge or dense rack deployments needing low power draw and high performance per watt.
Why Choose XS Network Tech?
We are a trusted partner for businesses of all sizes, including government departments, data centres, and service providers. Our expert team in sales, logistics, and customer service ensures that your needs are met with speed and precision, from pre-sales consultation to order fulfilment. With a commitment to fast delivery, sustainability, and a wide inventory, XS Network Tech is your go-to source for IT hardware. Trust us for quality, fast responses, and a hassle-free experience every time.