-
Vendor:Cisco
Cisco UCSX-GPU-L40 | NVIDIA L40 GPU, 48GB GDDR6, 300W, PCIe Gen4 x16
UCSX-GPU-L40
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-L40 | NVIDIA L40 GPU, 48GB GDDR6, 300W, PCIe Gen4 x16
The Cisco UCSX-GPU-L40 is a full-height, full-length dual-slot GPU accelerator designed for Cisco UCS X-Series modular systems. Built on the NVIDIA Ada Lovelace architecture with 48GB GDDR6 ECC memory and 18,176 CUDA cores, it delivers 90.5 TFLOPS FP32 performance for AI inference, virtual workstations, 3D rendering, and graphics-intensive workloads. The 300W passive-cooled card includes PCIe Gen4 x16 connectivity and supports NVIDIA vGPU for multi-tenant virtualization in enterprise data center environments. -
Vendor:Cisco
Cisco UCSX-GPU-MI210 | Instinct MI210 GPU, 64GB HBM2e, 300W, PCIe 4.0 x16
UCSX-GPU-MI210
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-MI210 | Instinct MI210 GPU, 64GB HBM2e, 300W, PCIe 4.0 x16
The Cisco UCSX-GPU-MI210 AMD Instinct MI210 accelerator delivers exceptional double-precision compute performance for HPC and AI workloads in enterprise data centers. Built on the AMD CDNA 2 architecture with 64GB HBM2e memory and 1.6 TB/s bandwidth, this passive-cooled PCIe Gen4 x16 GPU provides 45.3 TFLOPS FP64 matrix performance and 181 TFLOPS FP16/BF16 throughput. Designed for Cisco UCS X-Series modular systems, it supports multi-GPU scaling via AMD Infinity Fabric links for demanding scientific computing, deep learning, and simulation applications. -
Vendor:Cisco
Cisco UCSX-GPU-T4-16-D | NVIDIA T4 GPU, 16GB GDDR6, 70W, PCIe 3.0 x16, Passive
UCSX-GPU-T4-16-D
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-T4-16-D | NVIDIA T4 GPU, 16GB GDDR6, 70W, PCIe 3.0 x16, Passive
The Cisco UCSX-GPU-T4-16-D is an NVIDIA Tesla T4 Tensor Core GPU designed for AI inference, machine learning, and virtualized workloads in Cisco UCS X-Series servers. Built on the NVIDIA Turing architecture with 2,560 CUDA cores and 320 Tensor Cores, it delivers up to 130 TOPS of INT8 performance for deep learning inference. Its low-profile, single-slot, 70W passive-cooled design enables high-density GPU deployment in space-constrained enterprise data centers and cloud environments. -
Vendor:Cisco
Cisco UCSX-GPU-T4-MEZZ | UCS X-Series T4 GPU Module, 16GB GDDR6, 75W Mezzanine
UCSX-GPU-T4-MEZZ
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-T4-MEZZ | UCS X-Series T4 GPU Module, 16GB GDDR6, 75W Mezzanine
The Cisco UCSX-GPU-T4-MEZZ is a mezzanine GPU module designed for Cisco UCS X-Series compute nodes, featuring the NVIDIA Tesla T4 with 16GB GDDR6 memory and 2560 CUDA cores. Built on NVIDIA Turing architecture with 320 Tensor Cores, it delivers 8.1 TFLOPS FP32 and 130 TOPS INT8 performance for AI inference, machine learning, and video transcoding workloads. With 75W passive cooling and PCIe Gen3 x8 connectivity, it integrates directly into UCS X210c M6 servers for scalable GPU acceleration in enterprise data centers. -
Vendor:Cisco
Cisco UCSX-GPU-T4-16 | NVIDIA T4 Graphics Card, 16GB GDDR6, 75W, PCIe 3.0
UCSX-GPU-T4-16
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-T4-16 | NVIDIA T4 Graphics Card, 16GB GDDR6, 75W, PCIe 3.0
The Cisco UCSX-GPU-T4-16 features the NVIDIA T4 GPU with 16GB GDDR6 memory and NVIDIA Turing Tensor Cores, designed to accelerate AI inference, machine learning, deep learning training, data analytics, and virtual desktop workloads in enterprise data centers. Its energy-efficient 75W low-profile, single-slot design with passive cooling fits easily into mainstream rack servers and cloud infrastructures. Ideal for scale-out computing environments requiring GPU acceleration without high power or thermal overhead. -
Vendor:Cisco
Cisco UCSX-GPU-T4MEZZ-D | NVIDIA T4 Graphics Card, 16GB GDDR6, 75W, Mezzanine
UCSX-GPU-T4MEZZ-D
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-T4MEZZ-D | NVIDIA T4 Graphics Card, 16GB GDDR6, 75W, Mezzanine
The Cisco UCSX-GPU-T4MEZZ-D features an NVIDIA T4 Tensor Core GPU in a compact mezzanine form factor for Cisco UCS X-Series servers. With 16GB GDDR6 memory, 2560 CUDA cores, and 320 Tensor Cores, it accelerates AI inference, machine learning, virtual desktop infrastructure, and data analytics workloads. The passive cooling design and 75W TDP enable high-density GPU deployments in space-constrained data center environments. -
Vendor:Cisco
Cisco UCSX-X10C-GPUFM | X10c GPU front mezz | enables front-bay GPU interconnect | internal mezz interface | no external I/O
UCSX-X10C-GPUFM
Stock Status: Enquire
Vendor:CiscoCisco UCSX-X10C-GPUFM | X10c GPU front mezz | enables front-bay GPU interconnect | internal mezz interface | no external I/O
GPU Front Mezz module for the Cisco UCS X-Series X10c Compute Node. It provides the internal interconnect required to enable supported front-bay GPU options on the platform. Installs in the front mezz bay and does not present external I/O. -
Vendor:Cisco
Cisco UCSX-X10C-GPUFM-D | X10c GPU front mezz (variant) | front-bay GPU enablement | internal interconnect | no external I/O
UCSX-X10C-GPUFM-D
Stock Status: Enquire
Vendor:CiscoCisco UCSX-X10C-GPUFM-D | X10c GPU front mezz (variant) | front-bay GPU enablement | internal interconnect | no external I/O
GPU Front Mezz module variant for the Cisco UCS X-Series X10c Compute Node. Provides the internal interconnect needed for supported front-bay GPU options; installs in the front mezz bay with no external I/O exposure. -
Vendor:Cisco
Cisco UCSX-GPUA100-80-D | NVIDIA A100 80GB HBM2e GPU, 300W, PCIe Gen4, MIG
UCSX-GPUA100-80-D
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPUA100-80-D | NVIDIA A100 80GB HBM2e GPU, 300W, PCIe Gen4, MIG
The Cisco UCSX-GPUA100-80-D is an NVIDIA A100 Tensor Core GPU designed for enterprise AI training, HPC, and data analytics workloads in Cisco UCS X-Series modular systems. Featuring 80GB of HBM2e memory with up to 1.94 TB/s bandwidth, 300W TDP, and PCIe Gen4 x16 connectivity, this passively cooled graphics card delivers exceptional compute performance for demanding data center applications. Multi-Instance GPU technology enables partitioning into up to seven isolated instances, maximizing resource utilization for multi-tenant AI infrastructure. -
Vendor:NVIDIA
NVIDIA 900-52061-0000-100 | PCIe sync board for RTX, genlock/framelock, multi-GPU/multi-display synchronization
900-52061-0000-100
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-52061-0000-100 | PCIe sync board for RTX, genlock/framelock, multi-GPU/multi-display synchronization
Professional synchronization board that enables genlock and frame lock across multiple RTX GPUs and displays. Ideal for large visualization walls, broadcast graphics, simulation, and VR environments where precise timing cohesion is required. -
Vendor:NVIDIA
NVIDIA 900-5G153-2570-000 | RTX PRO 5000 72GB Blackwell Graphics Card, PCIe 5.0
900-5G153-2570-000
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-5G153-2570-000 | RTX PRO 5000 72GB Blackwell Graphics Card, PCIe 5.0
The NVIDIA RTX PRO 5000 72GB Blackwell is a professional workstation graphics card designed for AI development, data science, 3D rendering, and simulation workflows. Built on the Blackwell architecture with 72GB of GDDR7 ECC memory and 14,080 CUDA cores, it delivers 2,142 TOPS of AI performance and 1.344 TB/s memory bandwidth. This dual-slot GPU features PCIe 5.0 x16 interface, four DisplayPort 2.1b outputs supporting up to 7680x4320 resolution, and Multi-Instance GPU capability for parallel workloads. -
Vendor:NVIDIA
NVIDIA 920-9B020-00F0-0D1 | HDR 200Gb/s InfiniBand | Director-class chassis switch | Modular fabric core
920-9B020-00F0-0D1
Stock Status: Enquire
Vendor:NVIDIANVIDIA 920-9B020-00F0-0D1 | HDR 200Gb/s InfiniBand | Director-class chassis switch | Modular fabric core
Modular director-class chassis platform for HDR (200Gb/s) InfiniBand fabrics, engineered for large AI/HPC clusters requiring predictable latency and extreme bisectional bandwidth. Provides high-radix scalability, congestion control features, and adaptive routing to maintain performance at scale. Ideal as a non-blocking core for compute, storage, and GPU‑dense environments. -
Vendor:NVIDIA
NVIDIA 900-9X6AF-0016-ST2 | 1x QSFP28 100G InfiniBand EDR | PCIe 4.0 x8 | Low-profile | RDMA, GPUDirect, Adaptive Routing
900-9X6AF-0016-ST2
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X6AF-0016-ST2 | 1x QSFP28 100G InfiniBand EDR | PCIe 4.0 x8 | Low-profile | RDMA, GPUDirect, Adaptive Routing
High-throughput InfiniBand adapter for HPC and AI clusters, providing 100G EDR connectivity with ultra-low latency and advanced congestion control. QSFP28 interface supports short- and long-reach cabling options. PCIe 4.0 x8 host bandwidth and GPUDirect RDMA enable efficient, direct data paths for accelerated compute workflows. Built for dense deployments with a low-profile form factor. -
Vendor:NVIDIA
NVIDIA 900-9X6AF-0018-MT0 | 200Gb/s InfiniBand HDR, QSFP56, PCIe Gen4 x16, FHHL
900-9X6AF-0018-MT0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X6AF-0018-MT0 | 200Gb/s InfiniBand HDR, QSFP56, PCIe Gen4 x16, FHHL
High‑throughput InfiniBand adapter delivering HDR 200 Gb/s for HPC and AI clusters. Provides ultra‑low‑latency RDMA, GPUDirect RDMA, and in‑hardware offloads to accelerate MPI messaging, storage fabrics, and distributed training workloads. Ideal for scaling InfiniBand fabrics and enabling NVMe‑oF/IB in performance‑critical environments. -
Vendor:NVIDIA
NVIDIA 980-9I46J-00N01A | OSFP AOC, 400G InfiniBand NDR/400GbE, OM4 MMF 850nm, 1.5 m, OSFP–OSFP
980-9I46J-00N01A
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9I46J-00N01A | OSFP AOC, 400G InfiniBand NDR/400GbE, OM4 MMF 850nm, 1.5 m, OSFP–OSFP
High-performance active optical cable for short-reach InfiniBand NDR and 400GbE links. Integrated optics and fiber deliver consistent latency and signal integrity for leaf–spine, storage, and GPU fabric interconnects. -
Vendor:NVIDIA
NVIDIA 900-9X81E-00EX-SL0 | HHHL SuperNIC | 1x800Gb/s XDR InfiniBand (OSFP, single-cage) or 2x400GbE | PCIe Gen5 x16 | C8180L Partner Cooled
900-9X81E-00EX-SL0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X81E-00EX-SL0 | HHHL SuperNIC | 1x800Gb/s XDR InfiniBand (OSFP, single-cage) or 2x400GbE | PCIe Gen5 x16 | C8180L Partner Cooled
High-performance HHHL SuperNIC designed for accelerated AI/HPC and ultra‑low‑latency fabrics. A single OSFP cage supports 800Gb/s XDR InfiniBand by default, or can be configured for 2x400GbE to fit converged data center needs. Deliver deterministic latency with advanced RDMA/RoCEv2 offloads, deep virtualization (SR‑IOV), and precise time sync. Ideal for GPU‑dense servers and storage nodes requiring extreme throughput, congestion control, and scalable fabric connectivity.
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.