<h2 class="" data-path-to-node="3">NVIDIA RTX PRO 6000 Blackwell (96GB) – HPE Server Edition</h2><p data-path-to-node="4"><strong class="" data-path-to-node="4" data-index-in-node="0">Part Number: S6A73C</strong></p><p data-path-to-node="5"><span class="">Empower your enterprise with the pinnacle of visual computing and AI performance.</span><span class=""> The </span><strong class="" data-path-to-node="5" data-index-in-node="86">NVIDIA RTX PRO 6000 Blackwell Server Edition (S6A73C)</strong><span class=""> is engineered for the most demanding professional workloads,</span><span class=""> from massive-scale AI model fine-tuning to photorealistic cinematic rendering and complex engineering simulations.</span></p><p data-path-to-node="6"><span class="">With a staggering </span><strong class="" data-path-to-node="6" data-index-in-node="18">96 GB of GDDR7 memory</strong><span class=""> and the cutting-edge </span><strong class="" data-path-to-node="6" data-index-in-node="61">Blackwell architecture</strong><span class="">,</span><span class=""> this accelerator provides the memory capacity and computational throughput required to handle multi-billion-parameter datasets with ease.</span></p><h3 class="" data-path-to-node="7">Key Features</h3><ul data-path-to-node="8"><li><p data-path-to-node="8,0,0"><strong class="" data-path-to-node="8,0,0" data-index-in-node="0">Next-Gen Blackwell Architecture:</strong><span class=""> Features 5th Gen Tensor Cores and 4th Gen Ray Tracing (RT) Cores,</span><span class=""> delivering up to 3x the performance of previous generations for AI and rendering.</span></p></li><li><p data-path-to-node="8,1,0"><strong class="" data-path-to-node="8,1,0" data-index-in-node="0">Massive 96GB GDDR7 Memory:</strong><span class=""> Industry-leading capacity and speed (1.</span><span class="">8 TB/s bandwidth) allow for the local processing of massive 3D scenes and LLM (Large Language Model) fine-tuning.</span></p></li><li><p data-path-to-node="8,2,0"><strong class="" data-path-to-node="8,2,0" data-index-in-node="0">Server-Optimized Design:</strong><span class=""> Specifically designed for HPE ProLiant and Apollo servers,</span><span class=""> featuring a double-flow-through thermal design to maintain peak stability under 600W power loads.</span></p></li><li><p data-path-to-node="8,3,0"><strong class="" data-path-to-node="8,3,0" data-index-in-node="0">Advanced AI Precision:</strong><span class=""> Supports </span><strong class="" data-path-to-node="8,3,0" data-index-in-node="32">FP4 precision</strong><span class="">,</span><span class=""> enabling faster AI processing times with reduced memory overhead for generative AI applications.</span></p></li><li><p data-path-to-node="8,4,0"><strong class="" data-path-to-node="8,4,0" data-index-in-node="0">PCIe Gen 5 Ready:</strong><span class=""> Doubles the data-transfer speeds from the CPU compared to Gen 4,</span><span class=""> eliminating bottlenecks for data-intensive tasks.</span></p></li></ul><p><span class=""><br data-mce-bogus="1"></span></p><table data-path-to-node="11" class="mce-item-table"><thead><tr><td style="text-align: center;" data-mce-style="text-align: center;"><strong>Feature</strong></td><td style="text-align: center;" data-mce-style="text-align: center;"><strong>Specification</strong></td></tr></thead><tbody><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,1,0,0"><strong data-path-to-node="11,1,0,0" data-index-in-node="0">GPU Architecture</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,1,1,0">NVIDIA Blackwell</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,2,0,0"><strong data-path-to-node="11,2,0,0" data-index-in-node="0">Memory</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,2,1,0">96 GB GDDR7</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,3,0,0"><strong data-path-to-node="11,3,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,3,1,0">1,792 GB/s (approx. 1.8 TB/s)</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,4,0,0"><strong data-path-to-node="11,4,0,0" data-index-in-node="0">Interface</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,4,1,0">PCI Express 5.0 x16</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,5,0,0"><strong data-path-to-node="11,5,0,0" data-index-in-node="0">Max Power Consumption</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,5,1,0">600 W</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,6,0,0"><strong data-path-to-node="11,6,0,0" data-index-in-node="0">Thermal Solution</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,6,1,0">Passive/Server-Controlled (Double-Flow-Through)</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,7,0,0"><strong data-path-to-node="11,7,0,0" data-index-in-node="0">Display Connectors</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,7,1,0">4x DisplayPort 2.1b</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,8,0,0"><strong data-path-to-node="11,8,0,0" data-index-in-node="0">Multi-Instance GPU (MIG)</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,8,1,0">Supported (up to 7 instances)</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,9,0,0"><strong data-path-to-node="11,9,0,0" data-index-in-node="0">Ray Tracing Cores</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,9,1,0">4th Generation</span></td></tr><tr><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,10,0,0"><strong data-path-to-node="11,10,0,0" data-index-in-node="0">Tensor Cores</strong></span></td><td style="text-align: center;" data-mce-style="text-align: center;"><span data-path-to-node="11,10,1,0">5th Generation</span></td></tr></tbody></table>

India

<h3 data-path-to-node="0"><strong data-path-to-node="0" data-index-in-node="0">HPE NVIDIA RTX 6000 Blackwell 96GB PCIe Accelerator (S6A73C)</strong></h3><p data-path-to-node="1">Unlock the next generation of professional performance with the <strong data-path-to-node="1" data-index-in-node="64">HPE NVIDIA RTX™ 6000 Blackwell Server Edition</strong>. Built on the revolutionary Blackwell architecture, this flagship GPU features a massive <strong data-path-to-node="1" data-index-in-node="199">96GB of high-speed GDDR7 memory</strong> to tackle the world’s most complex AI, rendering, and data science workloads.</p><ul data-path-to-node="2"><li><p data-path-to-node="2,0,0"><strong data-path-to-node="2,0,0" data-index-in-node="0">Next-Gen Power:</strong> Features 5th Gen Tensor Cores and 4th Gen RT Cores for unparalleled AI and ray-tracing throughput.</p></li><li><p data-path-to-node="2,1,0"><strong data-path-to-node="2,1,0" data-index-in-node="0">Massive Memory:</strong> 96GB capacity handles multi-billion parameter AI models and enormous datasets with ease.</p></li><li><p data-path-to-node="2,2,0"><strong data-path-to-node="2,2,0" data-index-in-node="0">Enterprise Stability:</strong> Optimized for HPE ProLiant servers with a server-grade thermal design and PCIe Gen 5 support.</p></li><li><p data-path-to-node="2,3,0"><strong data-path-to-node="2,3,0" data-index-in-node="0">Ultimate Precision:</strong> Supports the latest FP4 and FP8 formats for lightning-fast generative AI training and inference.</p></li></ul><p data-path-to-node="3"><strong data-path-to-node="3" data-index-in-node="0">The definitive choice for AI fine-tuning, photorealistic digital twins, and high-performance computing (HPC).</strong></p>

NVIDIA RTX PRO 6000 Blackwell Server Edition 96GB PCIe Accelerator for HPE

NVIDIA

<p class="first-token" data-path-to-node="0"><span class="">The </span><strong class="" data-path-to-node="0" data-index-in-node="4">S2D86C</strong><span class=""> is the HPE (Hewlett Packard Enterprise) manufacturer part number for the </span><strong class="" data-path-to-node="0" data-index-in-node="84">NVIDIA H100 NVL 94GB PCIe Accelerator</strong><span class="">.</span><span class="">Based on the NVIDIA Hopper™ architecture,</span><span class=""> this is an enterprise-grade GPU specifically designed for the most demanding artificial intelligence (AI),</span><span class=""> high-performance computing (HPC),</span><span class=""> and large language model (LLM) workloads.</span></p><table data-path-to-node="7" class="mce-item-table" data-mce-selected="1"><thead><tr><td><strong>Feature</strong></td><td><strong>Specification</strong></td></tr></thead><tbody><tr><td><span data-path-to-node="7,1,0,0"><strong data-path-to-node="7,1,0,0" data-index-in-node="0">HPE Part Number</strong></span></td><td><span data-path-to-node="7,1,1,0">S2D86C</span></td></tr><tr><td><span data-path-to-node="7,2,0,0"><strong data-path-to-node="7,2,0,0" data-index-in-node="0">GPU Architecture</strong></span></td><td><span data-path-to-node="7,2,1,0">NVIDIA Hopper™</span></td></tr><tr><td><span data-path-to-node="7,3,0,0"><strong data-path-to-node="7,3,0,0" data-index-in-node="0">Memory Capacity</strong></span></td><td><span data-path-to-node="7,3,1,0">94 GB HBM3</span></td></tr><tr><td><span data-path-to-node="7,4,0,0"><strong data-path-to-node="7,4,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td><span data-path-to-node="7,4,1,0">3.9 TB/s</span></td></tr><tr><td><span data-path-to-node="7,5,0,0"><strong data-path-to-node="7,5,0,0" data-index-in-node="0">Interface</strong></span></td><td><span data-path-to-node="7,5,1,0">PCI Express 5.0 x16</span></td></tr><tr><td><span data-path-to-node="7,6,0,0"><strong data-path-to-node="7,6,0,0" data-index-in-node="0">Form Factor</strong></span></td><td><span data-path-to-node="7,6,1,0">Dual-Slot PCIe Card</span></td></tr><tr><td><span data-path-to-node="7,7,0,0"><strong data-path-to-node="7,7,0,0" data-index-in-node="0">TDP (Power)</strong></span></td><td><span data-path-to-node="7,7,1,0">Configurable (Default Max ~350W - 400W)</span></td></tr><tr><td><span data-path-to-node="7,8,0,0"><strong data-path-to-node="7,8,0,0" data-index-in-node="0">Cooling</strong></span></td><td><span data-path-to-node="7,8,1,0">Passive (requires server-directed airflow)</span></td></tr></tbody></table><div id="mceResizeHandlenw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;" style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;"></div><div id="mceResizeHandlene" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;" style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;"></div><div id="mceResizeHandlese" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 529.545px;" style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 529.545px;"></div><div id="mceResizeHandlesw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 529.545px;" style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 529.545px;"></div>

<p data-path-to-node="0">The <strong data-path-to-node="0" data-index-in-node="4">HPE S2D86C</strong> is a high-performance <strong data-path-to-node="0" data-index-in-node="37">NVIDIA H100 NVL 94GB PCIe Accelerator</strong> designed for the most intensive AI and HPC workloads.Built on the <strong data-path-to-node="1" data-index-in-node="13">NVIDIA Hopper architecture</strong>, it features 94GB of ultra-fast HBM3 memory and a dedicated Transformer Engine to accelerate Large Language Model (LLM) training and inference. As a PCIe 5.0 card, it is engineered for seamless integration into <strong data-path-to-node="1" data-index-in-node="251">HPE ProLiant Gen11/Gen12</strong> servers, providing the massive memory bandwidth and compute power required for generative AI, data analytics, and complex scientific simulations.</p>

NVIDIA H100 NVL 94GB PCIe Accelerator

<p class="first-token" data-path-to-node="0"><span class="">The </span><strong class="" data-path-to-node="0" data-index-in-node="4">S3U30C</strong><span class=""> is the HPE (Hewlett Packard Enterprise) product part number for the </span><strong class="" data-path-to-node="0" data-index-in-node="79">NVIDIA H200 NVL 141GB PCIe Accelerator</strong><span class="">.</span><span class="">This is a professional-grade GPU based on the NVIDIA Hopper architecture,</span><span class=""> specifically designed for high-performance computing (HPC) and generative AI workloads.</span><span class=""> It is the first GPU to utilize </span><strong class="" data-path-to-node="1" data-index-in-node="193">HBM3e memory</strong><span class="">,</span><span class=""> offering a significant jump in memory bandwidth and capacity over the previous H100 generation.</span></p><table data-path-to-node="4" class="mce-item-table" data-mce-selected="1"><thead><tr><td><strong>Feature</strong></td><td><strong>Specification</strong></td></tr></thead><tbody><tr><td><span data-path-to-node="4,1,0,0"><strong data-path-to-node="4,1,0,0" data-index-in-node="0">GPU Architecture</strong></span></td><td><span data-path-to-node="4,1,1,0">NVIDIA Hopper</span></td></tr><tr><td><span data-path-to-node="4,2,0,0"><strong data-path-to-node="4,2,0,0" data-index-in-node="0">Memory Capacity</strong></span></td><td><span data-path-to-node="4,2,1,0">141 GB HBM3e</span></td></tr><tr><td><span data-path-to-node="4,3,0,0"><strong data-path-to-node="4,3,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td><span data-path-to-node="4,3,1,0">4.8 TB/s</span></td></tr><tr><td><span data-path-to-node="4,4,0,0"><strong data-path-to-node="4,4,0,0" data-index-in-node="0">Performance (FP64)</strong></span></td><td><span data-path-to-node="4,4,1,0">34 TFLOPS (Peak Double Precision)</span></td></tr><tr><td><span data-path-to-node="4,5,0,0"><strong data-path-to-node="4,5,0,0" data-index-in-node="0">Performance (FP32)</strong></span></td><td><span data-path-to-node="4,5,1,0">67 TFLOPS (Peak Single Precision)</span></td></tr><tr><td><span data-path-to-node="4,6,0,0"><strong data-path-to-node="4,6,0,0" data-index-in-node="0">Performance (FP16)</strong></span></td><td><span data-path-to-node="4,6,1,0">1,979 TFLOPS (Peak Half Precision)</span></td></tr><tr><td><span data-path-to-node="4,7,0,0"><strong data-path-to-node="4,7,0,0" data-index-in-node="0">Interface</strong></span></td><td><span data-path-to-node="4,7,1,0">PCI Express 5.0 x16</span></td></tr><tr><td><span data-path-to-node="4,8,0,0"><strong data-path-to-node="4,8,0,0" data-index-in-node="0">Thermal Design Power (TDP)</strong></span></td><td><span data-path-to-node="4,8,1,0">600W</span></td></tr><tr><td><span data-path-to-node="4,9,0,0"><strong data-path-to-node="4,9,0,0" data-index-in-node="0">Form Factor</strong></span></td><td><span data-path-to-node="4,9,1,0">Full-Height, Full-Length (FHFL)</span></td></tr></tbody></table><div id="mceResizeHandlenw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;" style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;"></div><div id="mceResizeHandlene" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;" style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;"></div><div id="mceResizeHandlese" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 577.727px;" style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 577.727px;"></div><div id="mceResizeHandlesw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 577.727px;" style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 577.727px;"></div>

<p class="first-token" data-path-to-node="0"><span class="">The </span><strong class="" data-path-to-node="0" data-index-in-node="4">S3U30C</strong><span class=""> is the HPE-specific part number for the </span><strong class="" data-path-to-node="0" data-index-in-node="51">NVIDIA H200 NVL 141GB PCIe Accelerator</strong><span class="">.</span><span class="">It is a high-performance GPU designed for enterprise servers,</span><span class=""> specifically optimized for large-scale AI inference,</span><span class=""> generative AI,</span><span class=""> and high-performance computing (HPC) workloads.</span></p>

NVIDIA H200 NVL 141GB PCIe Accelerator

<p class="first-token" data-path-to-node="0"><span class="">The </span><strong class="" data-path-to-node="0" data-index-in-node="4">R9W59C</strong><span class=""> is a high-performance GPU accelerator manufactured by </span><strong class="" data-path-to-node="0" data-index-in-node="65">HPE (Hewlett Packard Enterprise)</strong><span class="">,</span><span class=""> specifically the </span><strong class="" data-path-to-node="0" data-index-in-node="116">NVIDIA A10 24GB PCIe Non-CEC Accelerator</strong><span class="">.</span><span class="">While often listed under "HPE Aruba Networking" in some vendor catalogs because of HPE's unified branding,</span><span class=""> this is primarily a data center GPU designed for </span><strong class="" data-path-to-node="1" data-index-in-node="156">HPE ProLiant</strong><span class=""> and </span><strong class="" data-path-to-node="1" data-index-in-node="173">Synergy</strong><span class=""> servers to handle AI,</span><span class=""> virtualization,</span><span class=""> and high-end graphics.</span></p><table data-path-to-node="3" class="mce-item-table" data-mce-selected="1"><thead><tr><td><strong>Feature</strong></td><td><strong>Specification</strong></td></tr></thead><tbody><tr><td><span data-path-to-node="3,1,0,0"><strong data-path-to-node="3,1,0,0" data-index-in-node="0">GPU Model</strong></span></td><td><span data-path-to-node="3,1,1,0">NVIDIA A10</span></td></tr><tr><td><span data-path-to-node="3,2,0,0"><strong data-path-to-node="3,2,0,0" data-index-in-node="0">Architecture</strong></span></td><td><span data-path-to-node="3,2,1,0">NVIDIA Ampere</span></td></tr><tr><td><span data-path-to-node="3,3,0,0"><strong data-path-to-node="3,3,0,0" data-index-in-node="0">Memory</strong></span></td><td><span data-path-to-node="3,3,1,0">24GB GDDR6</span></td></tr><tr><td><span data-path-to-node="3,4,0,0"><strong data-path-to-node="3,4,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td><span data-path-to-node="3,4,1,0">600 GB/s</span></td></tr><tr><td><span data-path-to-node="3,5,0,0"><strong data-path-to-node="3,5,0,0" data-index-in-node="0">Interface</strong></span></td><td><span data-path-to-node="3,5,1,0">PCIe 4.0 x16</span></td></tr><tr><td><span data-path-to-node="3,6,0,0"><strong data-path-to-node="3,6,0,0" data-index-in-node="0">Power Consumption</strong></span></td><td><span data-path-to-node="3,6,1,0">150W (TDP)</span></td></tr><tr><td><span data-path-to-node="3,7,0,0"><strong data-path-to-node="3,7,0,0" data-index-in-node="0">Form Factor</strong></span></td><td><span data-path-to-node="3,7,1,0">Single-Slot, Full-Height</span></td></tr><tr><td><span data-path-to-node="3,8,0,0"><strong data-path-to-node="3,8,0,0" data-index-in-node="0">Cooling</strong></span></td><td><span data-path-to-node="3,8,1,0">Passive (requires server airflow)</span></td></tr></tbody></table><div id="mceResizeHandlenw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;" style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;"></div><div id="mceResizeHandlene" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;" style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;"></div><div id="mceResizeHandlese" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 529.545px;" style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 529.545px;"></div><div id="mceResizeHandlesw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 529.545px;" style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 529.545px;"></div>

<p data-path-to-node="0">The <strong data-path-to-node="0" data-index-in-node="4">HPE R9W59C</strong> is a data center-grade <strong data-path-to-node="0" data-index-in-node="38">NVIDIA A10 24GB PCIe Non-CEC Accelerator</strong>. It is a single-slot, passively cooled GPU based on the NVIDIA Ampere architecture, specifically designed to accelerate mainstream AI inference, graphics rendering, and Virtual Desktop Infrastructure (VDI) workloads in HPE ProLiant and Synergy servers.</p><h3 data-path-to-node="1"><br data-mce-bogus="1"></h3>

NVIDIA A10 24GB PCIe NonCEC Accelerator

<p class="first-token" data-path-to-node="0"><span class="">The </span><strong class="" data-path-to-node="0" data-index-in-node="4">S0K89C</strong><span class=""> is the manufacturer part number for the </span><strong class="" data-path-to-node="0" data-index-in-node="51">NVIDIA L4 24GB PCIe Accelerator for HPE (Hewlett Packard Enterprise)</strong><span class="">.</span><span class="">This is a professional-grade GPU designed specifically for enterprise servers to handle high-demand workloads like Artificial Intelligence (AI) inference,</span><span class=""> video processing,</span><span class=""> and virtual workstations.</span></p><table data-path-to-node="4" class="mce-item-table" data-mce-selected="1"><thead><tr><td><strong>Feature</strong></td><td><strong>Details</strong></td></tr></thead><tbody><tr><td><span data-path-to-node="4,1,0,0"><strong data-path-to-node="4,1,0,0" data-index-in-node="0">GPU Architecture</strong></span></td><td><span data-path-to-node="4,1,1,0">NVIDIA Ada Lovelace (L4 Tensor Core)</span></td></tr><tr><td><span data-path-to-node="4,2,0,0"><strong data-path-to-node="4,2,0,0" data-index-in-node="0">Video Memory</strong></span></td><td><span data-path-to-node="4,2,1,0">24 GB GDDR6</span></td></tr><tr><td><span data-path-to-node="4,3,0,0"><strong data-path-to-node="4,3,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td><span data-path-to-node="4,3,1,0">300 GB/s</span></td></tr><tr><td><span data-path-to-node="4,4,0,0"><strong data-path-to-node="4,4,0,0" data-index-in-node="0">Interface</strong></span></td><td><span data-path-to-node="4,4,1,0">PCIe 4.0 x16 (some listings mention PCIe 5.0 compatibility)</span></td></tr><tr><td><span data-path-to-node="4,5,0,0"><strong data-path-to-node="4,5,0,0" data-index-in-node="0">Form Factor</strong></span></td><td><span data-path-to-node="4,5,1,0">Low Profile (Half-height, Half-length), Single Slot</span></td></tr><tr><td><span data-path-to-node="4,6,0,0"><strong data-path-to-node="4,6,0,0" data-index-in-node="0">Cooling</strong></span></td><td><span data-path-to-node="4,6,1,0">Passive (requires server airflow)</span></td></tr><tr><td><span data-path-to-node="4,7,0,0"><strong data-path-to-node="4,7,0,0" data-index-in-node="0">Power Consumption</strong></span></td><td><span data-path-to-node="4,7,1,0">72W (Max TDP)</span></td></tr></tbody></table><div id="mceResizeHandlenw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 77.7273px;" style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 77.7273px;"></div><div id="mceResizeHandlene" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 77.7273px;" style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 77.7273px;"></div><div id="mceResizeHandlese" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 465px;" style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 465px;"></div><div id="mceResizeHandlesw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 465px;" style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 465px;"></div>

<div class="user-query-container"><div id="user-query-content-1" class="query-content ng-star-inserted verticle-align-for-single-line-text" data-hveid="0" data-ved="0CAAQ3ucQahgKEwiqqZfD8fiRAxUAAAAAHQAAAAAQowQ"><div class="query-text gds-body-l" dir="ltr" role="heading" aria-level="2"></div></div></div><div><div class="response-container ng-tns-c199654419-99 response-container-with-gpi ng-star-inserted response-container-has-multiple-responses" data-hveid="1"><div class="response-container-header ng-tns-c199654419-99 ng-star-inserted"><div class="response-container-header-controls ng-tns-c199654419-99"><div class="response-tts-container ng-star-inserted" data-test-draft-id="rc_5ccb600f261099f0"></div></div><div class="response-container-header-status ng-tns-c199654419-99"></div></div><div class="presented-response-container ng-tns-c199654419-99" data-hveid="2"><div class="avatar-gutter ng-tns-c199654419-99 ng-star-inserted"><div class="bard-avatar ng-tns-c4035847673-100"><div class="avatar-container ng-tns-c4035847673-100"><div class="avatar avatar_primary ng-tns-c4035847673-100 ng-star-inserted"><div class="avatar_primary_model is-gpi-avatar ng-tns-c4035847673-100"></div></div></div></div></div><div class="response-container-content ng-tns-c199654419-99"><div class="response-content ng-tns-c199654419-99"><div class="container"><div id="model-response-message-contentr_7f8bc7d5709745f6" class="markdown markdown-main-panel stronger enable-updated-hr-color" dir="ltr" aria-live="polite" aria-busy="false"><p data-path-to-node="0">The&nbsp;<strong data-path-to-node="0" data-index-in-node="4">HPE S0K89C</strong> is an enterprise-grade NVIDIA L4 Tensor Core GPU designed for mainstream servers. It is a compact, energy-efficient accelerator optimized for AI inference, video processing, and virtual workstations.</p></div></div></div></div></div></div></div>

NVIDIA L4 24GB PCIe Accelerator

<p><span class="">The NVIDIA RTX 4000 Ada is built on the highly efficient </span><strong class="" data-path-to-node="3" data-index-in-node="57">Ada Lovelace architecture</strong><span class="">.</span><span class=""> It strikes a balance between power and performance,</span><span class=""> offering significant leaps in Ray Tracing (RT) and Tensor Core performance compared to previous generations.</span><span class=""> For enterprise users,</span><span class=""> this means faster photorealistic rendering,</span><span class=""> enhanced AI-driven graphics,</span><span class=""> and more efficient data processing.</span></p><table data-path-to-node="8" class="mce-item-table" data-mce-selected="1"><thead><tr><td><strong>Feature</strong></td><td><strong>Specification</strong></td></tr></thead><tbody><tr><td><span data-path-to-node="8,1,0,0"><strong data-path-to-node="8,1,0,0" data-index-in-node="0">GPU Architecture</strong></span></td><td><span data-path-to-node="8,1,1,0">NVIDIA Ada Lovelace</span></td></tr><tr><td><span data-path-to-node="8,2,0,0"><strong data-path-to-node="8,2,0,0" data-index-in-node="0">Memory</strong></span></td><td><span data-path-to-node="8,2,1,0">20 GB GDDR6</span></td></tr><tr><td><span data-path-to-node="8,3,0,0"><strong data-path-to-node="8,3,0,0" data-index-in-node="0">Memory Bandwidth</strong></span></td><td><span data-path-to-node="8,3,1,0">Up to 360 GB/s</span></td></tr><tr><td><span data-path-to-node="8,4,0,0"><strong data-path-to-node="8,4,0,0" data-index-in-node="0">Memory Interface</strong></span></td><td><span data-path-to-node="8,4,1,0">160-bit</span></td></tr><tr><td><span data-path-to-node="8,5,0,0"><strong data-path-to-node="8,5,0,0" data-index-in-node="0">CUDA Cores</strong></span></td><td><span data-path-to-node="8,5,1,0">6,144</span></td></tr><tr><td><span data-path-to-node="8,6,0,0"><strong data-path-to-node="8,6,0,0" data-index-in-node="0">Interface</strong></span></td><td><span data-path-to-node="8,6,1,0">PCI Express 4.0 x16</span></td></tr><tr><td><span data-path-to-node="8,7,0,0"><strong data-path-to-node="8,7,0,0" data-index-in-node="0">Max Power Consumption</strong></span></td><td><span data-path-to-node="8,7,1,0">130 W</span></td></tr><tr><td><span data-path-to-node="8,8,0,0"><strong data-path-to-node="8,8,0,0" data-index-in-node="0">Cooling</strong></span></td><td><span data-path-to-node="8,8,1,0">Active (Blower Fan)</span></td></tr><tr><td><span data-path-to-node="8,9,0,0"><strong data-path-to-node="8,9,0,0" data-index-in-node="0">Display Connectors</strong></span></td><td><span data-path-to-node="8,9,1,0">4x DisplayPort 1.4a</span></td></tr><tr><td><span data-path-to-node="8,10,0,0"><strong data-path-to-node="8,10,0,0" data-index-in-node="0">Form Factor</strong></span></td><td><span data-path-to-node="8,10,1,0">Single-Slot, Full-Height</span></td></tr></tbody></table><div id="mceResizeHandlenw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;" style="cursor: nw-resize; margin: 0px; padding: 0px; left: 20px; top: 94.0909px;"></div><div id="mceResizeHandlene" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;" style="cursor: ne-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 94.0909px;"></div><div id="mceResizeHandlese" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 625.909px;" style="cursor: se-resize; margin: 0px; padding: 0px; left: 1040.91px; top: 625.909px;"></div><div id="mceResizeHandlesw" data-mce-bogus="all" class="mce-resizehandle" unselectable="true" data-mce-style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 625.909px;" style="cursor: sw-resize; margin: 0px; padding: 0px; left: 20px; top: 625.909px;"></div>

<p data-path-to-node="0">The <strong data-path-to-node="0" data-index-in-node="4">HPE S3T54C</strong> is an enterprise-grade <strong data-path-to-node="0" data-index-in-node="38">NVIDIA RTX 4000 Ada Generation</strong> Graphics Accelerator with <strong data-path-to-node="0" data-index-in-node="95">20GB GDDR6</strong> memory.Built on the high-efficiency Ada Lovelace architecture, this single-slot GPU is designed for <strong data-path-to-node="1" data-index-in-node="93">HPE ProLiant Gen11</strong> servers to accelerate professional workloads such as AI inferencing, 3D rendering, and complex CAD applications. It delivers a significant boost in performance-per-watt, making it ideal for workstations and servers where power efficiency and high-density compute are required.</p>

NVIDIA RTX 4000 Ada GPU Module

<h1 class="pagetitle" data-mce-tabindex="0"><strong>HPE NVIDIA L40S PCIe 4.0 x16 Graphics Card, 48GB GDDR6</strong></h1><p>Do you require higher performance for artificial intelligence (AI) training and inference, high-performance computing (HPC), or graphics? NVIDIA Accelerators for HPE help solve the world’s most important scientific, industrial, and business challenges with AI and HPC. Visualize complex <a data-mce-href="https://www.kcomputers.com" href="https://www.kcomputers.com">content</a> to create cutting-edge products, tell immersive stories, and reimagine cities of the future. Extract new insights from massive datasets. Hewlett Packard Enterprise servers with NVIDIA accelerators are designed for the age of elastic computing, providing unmatched acceleration at every scale.</p><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">PCIe Express Gen5 provides increased bandwidth and improves data-transfer speeds from CPU memory.</div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><br data-mce-bogus="1"></div></div></div><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">Fourth-generation tensor cores for dramatic AI speedups.</div></div></div><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><br data-mce-bogus="1"></div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">Faster GPU memory to boost performance.</div></div></div><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><br data-mce-bogus="1"></div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">Third-generation NVLink doubles the GPU-GPU direct bandwidth.</div></div></div><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><br data-mce-bogus="1"></div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">Third-generation RT cores for <a data-mce-href="https://www.kcomputers.com" href="https://www.kcomputers.com">speeding</a> up rendering workloads.</div></div></div><div class="ccs-cc-inline-features-block ccs-cc-inline-single-feature"><div class="ccs-cc-inline-feature" data-type="text"><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><br data-mce-bogus="1"></div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description">NVIDIA confidential computing is a built-in security feature of Hopper-based accelerators.</div><div class="ccs-cc-inline-feature-content ccs-cc-inline-feature-description"><h1 class="pagetitle" data-mce-tabindex="0"><strong>Specifications:</strong></h1><p><strong>Memory size per board: 48 GB</strong></p><p><strong>Product dimensions (imperial): 1.37 x 10.51 x 4.37 in</strong></p><p><strong>Weight: <a data-mce-href="https://www.kcomputers.com" href="https://www.kcomputers.com">3.743 lb﻿</a></strong></p><p><br data-mce-bogus="1"></p><p><br data-mce-bogus="1"></p><p><br data-mce-bogus="1"></p></div></div></div>

<p>Build new AI models with supervised or unsupervised training for generative AI, computer vision, large language models, scientific discovery, and financial market modeling with NVIDIA accelerators and HPE Cray systems.</p>

HPE NVIDIA L40S PCIe 4.0 x16 Graphics Card, 48GB GDDR6

Graphic Cards

Maximize performance with HPE GPU servers. Engineered for AI training, inference, and high-performance computing (HPC). Deploy scalable GPU power today.