[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-1-en-105":3,"doc-seo-563630-105":53,"doc-detail-563630-en":126},{"code":4,"msg":5,"data":6},0,"success",[7,14,19,24,29,34,39,44,49],{"id":8,"doc_module":9,"doc_module_name":10,"category_name":11,"show_sort_weight":12,"slug":13},11,1,"Template","Presentations",90,"presentations",{"id":15,"doc_module":9,"doc_module_name":10,"category_name":16,"show_sort_weight":17,"slug":18},12,"Resumes",80,"resumes",{"id":20,"doc_module":9,"doc_module_name":10,"category_name":21,"show_sort_weight":22,"slug":23},14,"Invoices",70,"invoices",{"id":25,"doc_module":9,"doc_module_name":10,"category_name":26,"show_sort_weight":27,"slug":28},15,"Posters",60,"posters",{"id":30,"doc_module":9,"doc_module_name":10,"category_name":31,"show_sort_weight":32,"slug":33},16,"Social Media",50,"social-media",{"id":35,"doc_module":9,"doc_module_name":10,"category_name":36,"show_sort_weight":37,"slug":38},17,"Forms",40,"forms",{"id":40,"doc_module":9,"doc_module_name":10,"category_name":41,"show_sort_weight":42,"slug":43},18,"Letters",30,"letters",{"id":45,"doc_module":9,"doc_module_name":10,"category_name":46,"show_sort_weight":47,"slug":48},21,"Paper Templates",5,"papers-templates",{"id":50,"doc_module":9,"doc_module_name":10,"category_name":51,"show_sort_weight":4,"slug":52},158,"General","general-158",{"code":4,"msg":54,"data":55},"ok",{"site_id":56,"language":57,"slug":58,"title":59,"keywords":60,"description":61,"schema_data":62,"social_meta":119,"head_meta":121,"extra_data":123,"updated_unix":125},105,"en","line-card-nvidia-data-center-platform","Line Card - NVIDIA Data Center Platform","","Line card for the NVIDIA Data Center Platform, outlining how advanced AI reasoning workloads benefit from accelerated computing across pretraining, posttraining, and test-time scaling. It links each scaling approach to distinct data center infrastructure requirements and presents a unified portfolio for AI reasoning, simulation, digital twins, and advanced graphics. The guide helps evaluate GPU and solution selection for performance and innovation, including solution categories, switching and networking options, and pricing-performance guidance based on measured workloads.",{"@graph":63,"@context":118},[64,80,101],{"@type":65,"itemListElement":66},"BreadcrumbList",[67,71,74,77],{"item":68,"name":69,"@type":70,"position":9},"https://docshare.wps.com","Home","ListItem",{"item":72,"name":10,"@type":70,"position":73},"https://docshare.wps.com/template/",2,{"item":75,"name":11,"@type":70,"position":76},"https://docshare.wps.com/template/presentations/",3,{"item":78,"name":59,"@type":70,"position":79},"https://docshare.wps.com/template/line-card-nvidia-data-center-platform/563630/",4,{"url":78,"name":59,"@type":81,"image":82,"author":87,"headline":59,"publisher":90,"fileFormat":93,"inLanguage":57,"description":61,"dateModified":94,"datePublished":95,"encodingFormat":93,"isAccessibleForFree":96,"interactionStatistic":97},"DigitalDocument",{"url":83,"@type":84,"width":85,"height":86},"https://docshare.wps.com/thumbnails/line-card-nvidia-data-center-platform/563630.png","ImageObject",442,249,{"name":88,"@type":89},"Sarah ","Person",{"url":68,"name":91,"@type":92},"DocShare","Organization","application/pdf","2026-10-19","2026-10-12",true,{"@type":98,"interactionType":99,"userInteractionCount":73},"InteractionCounter",{"@type":100},"ViewAction",{"@type":102,"mainEntity":103},"FAQPage",[104,110,114],{"name":105,"@type":106,"acceptedAnswer":107},"What scaling laws does the line card highlight for AI workloads?","Question",{"text":108,"@type":109},"It highlights three scaling laws: pretraining scaling, posttraining scaling, and test-time scaling during inference to enable reasoning.","Answer",{"name":111,"@type":106,"acceptedAnswer":112},"How does the document suggest using the price-performance comparisons?",{"text":113,"@type":109},"It states comparisons are relative within each column and should be used together with measured data for targeted workloads.",{"name":115,"@type":106,"acceptedAnswer":116},"What types of solutions does the NVIDIA Data Center Platform line card cover?",{"text":117,"@type":109},"It covers compute and networking configurations for AI reasoning, simulation, digital twins, advanced graphics, and related data center workload categories.","https://schema.org",{"og:url":78,"og:type":120,"og:title":59,"og:site_name":91,"og:description":61},"article",{"robots":122,"canonical":78},"index,follow",{"doc_id":124,"site_id":56},563630,1792434077,{"code":4,"msg":5,"data":127},{"doc_id":124,"user_id":128,"nickname":88,"user_avatar":129,"doc_module":9,"category_id":8,"category_name":11,"doc_title":59,"doc_description":61,"doc_content":130,"file_id":131,"file_url":132,"file_type":133,"file_size":134,"view_count":73,"is_deleted":4,"is_public":9,"is_downloadable":9,"audit_status":9,"page_count":76,"language":135,"language_code":57,"site_id":56,"html_lang":57,"table_of_contents":136,"faqs":137,"seo_title":138,"seo_description":61,"update_tm":139,"read_time":9},962085320529,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","Line Card  \nNVIDIA Data Center Platform  \nArtificial intelligence is driving a new era of advanced reasoning, revolutionizing workflows and transforming every industry. Accelerated computing allows organizations to address AI, scientific computing, digital twin, and graphics-based workloads. Meeting these challenges requires attention to the three scaling laws: pretraining scaling (initial large dataset training), posttraining scaling (enhancing models through fine-tuning), and test-time scaling (allocating more compute during inference to enable reasoning) . Each scaling law places unique demands on data center infrastructure.  \nThe NVIDIA accelerated computing platform provides a unified solution, delivering breakthrough performance for AI reasoning, simulation, digital twins, and advanced graphics. As AI workloads grow in complexity and diversity, selecting the right GPU—matched to your needs and specific use case—is critical for maximizing performance and innovation. This guide will help you navigate the NVIDIA solution portfolio, ensuring your infrastructure is ready for the next generation of AI reasoning, simulation, or graphics-driven applications.  \nNVIDIA Data Center Platform | Linecard | 1  \nNVIDIA Data Center Compute Portfolio  \n| Solution | \u003Cbr>\u003Cbr>Networking | \u003Cbr>Pretraining | \u003Cbr>Post-Training | \u003Cbr>\u003Cbr>Test-Time/Inference | \u003Cbr>Data Analytics | \u003Cbr>HPC/Scientific | \u003Cbr>AI Video Generation |\n| --- | --- | --- | --- | --- | --- | --- | --- |\n| GB300 NVL72\u003Cbr>GB200 NVL72\u003Cbr>GB200 NVL4\u003Cbr>GH200 NVL2 / NVL4\u003Cbr>HGX B300\u003Cbr>HGX B200\u003Cbr>HGX H200\u003Cbr>H200 NVL | QTM-X800/SPTMX\u003Cbr>QTM-X800/SPTMX\u003Cbr>QTM-2/SPTMX\u003Cbr>QTM-2/SPTMX\u003Cbr>QTM-X800/SPTMX\u003Cbr>QTM-X800/SPTMX\u003Cbr>QTM-2/SPTMX\u003Cbr>QTM-2/SPTMX | | | | | | |\n\nPrice-performance comparisons are relative within each column. This chart should be used in conjunction with measured data for targeted workloads.  \n Best  \nGood  \nQTM-X800 NVIDIA Quantum-X800 InfiniBand switch plus BlueField®-3 DPUs and ConnectX®-8 SuperNICs QTM-2 NVIDIA Quantum-2 InfiniBand switch plus BlueField-3 DPUs and ConnectX-7 NICs  \nSPTMX NVIDIA Spectrum-X™ Ethernet switch and SuperNICs plus BlueField-3 DPUs  \nNVIDIA Data Center Platform | Linecard | 2  \nNVIDIA Data Center Universal Graphics and Compute Portfolio  \n\n| Solution | \u003Cbr>\u003Cbr>Networking | \u003Cbr>Pretraining | \u003Cbr>Post-Training | \u003Cbr>\u003Cbr>Test-Time/ Inference | \u003Cbr>Data Analytics | \u003Cbr>HPC/Scientific | \u003Cbr>Omniverse™/\u003Cbr>Render | \u003Cbr>Virtualization1 vWS/VD | \u003Cbr>Far Edge/AI Video |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |\n| RTX PRO 6000\u003Cbr>Blackwell Server Edition\u003Cbr>L40S\u003Cbr>L4 | QTM-2/SPTMX\u003Cbr>QTM-2/SPTM3\u003Cbr>QTM-2/SPTM3 | | | | | | | | |\n\n1. vGPU is supported across all products shown  \nPrice-performance comparisons are relative within each column. This chart should be used in conjunction with measured data for targeted workloads.  \n Best  \nGood  \nQTM-2 NVIDIA Quantum-2 InfiniBand switch plus BlueField-3 DPUs and ConnectX-7 NICs SPTMX NVIDIA Spectrum-X Ethernet switch and SuperNICs plus BlueField-3 DPUSPTM3 NVIDIA Spectrum-3 Ethernet switch plus ConnectX-6 NICs and BlueField-2 DPUs  \nReady to Get Started?  \nFor more information on NVIDIA data center GPUs, visit: [nvidia.com/data-center-gpus](nvidia.com/data-center-gpus)  \n© 2025 NVIDIA Corporation and affiliates. All rights reserved. NVIDIA, the NVIDIA logo, BlueField, ConnectX, Omniverse, and SuperNIC are trademarks and/or registered trademarks of NVIDIA Corporation and affiliatesin the U.S. and other countries. Other company and product names may be trademarks of the respective owners with which they are associated. 4148700. AUG25","cbCaigRZvF9T3TUx","https://ap.wps.com/l/cbCaigRZvF9T3TUx","pdf",210075,"English","# NVIDIA Data Center Platform\n## Data Center Compute Portfolio\n## Data Center Universal Graphics and Compute Portfolio\n## Getting Started","[{\"question\":\"What scaling laws does the line card highlight for AI workloads?\",\"answer\":\"It highlights three scaling laws: pretraining scaling, posttraining scaling, and test-time scaling during inference to enable reasoning.\"},{\"question\":\"How does the document suggest using the price-performance comparisons?\",\"answer\":\"It states comparisons are relative within each column and should be used together with measured data for targeted workloads.\"},{\"question\":\"What types of solutions does the NVIDIA Data Center Platform line card cover?\",\"answer\":\"It covers compute and networking configurations for AI reasoning, simulation, digital twins, advanced graphics, and related data center workload categories.\"}]","Line Card - NVIDIA Data Center Platform | PDF",1791836906]