Performance-optimised single-socket AMD EPYC 4th generation based servers with up to four double-slot NVIDIA cards and a wide range of configuration options in space saving 2U form factor. “AI server 8 GPU, 2x AMD EPYC 9455 48C 300W 3. 15GHz, 384GB DDR5 4800MHz (24x 16GB), 3. 84 TB NVMe SSD, 4x. Your Deep Learning server, configurable with 4 high end deep learning GPUs which give you the fastest deep learning power available: upto 500 Trillion Tensor FLOPS of AI performance and 64 GB high speed GPU memory. Built to perform 24/7 at your inhouse data center or co-location with an server. Powerful and cost-efficient servers for AI workload. Available everywhere and at any time. Easy to use DNS management platform. List, add, modify or remove zones and records Our GEX-line is powered by NVIDIA GPUs with CUDA technology and is perfect for AI workloads and. Parallel computing is enabled with accelerators from NVIDIA, AMD, Intel, and others in GPU servers. Pre-installed with AI/ML software stack (PyTorch, TensorFlow, CUDA). Powered by the latest NVIDIA Blackwell architecture, AMD. KR4268V2 with AMD EPYC™ 9004 processors is a powerful 4U AI server featuring the latest PCIe Gen 5. 0 architecture and supporting up to 8 FHFL double-width GPUs. Its 10-card pass-through design significantly reduces CPU and GPU communication latency. This system is an ideal solution for advanced AI.