{"product_id":"nvidia-dgx-b200","title":"NVIDIA DGX B200 AI System, 8x NVIDIA Blackwell B200 GPUs (1,440GB HBM3e), 2x Intel Xeon Platinum 8570, 2TB RAM, 400Gb\/s ConnectX-7, (DGX-B200)","description":"\u003cdiv\u003e\n\u003cp style=\"font-family: Arial, sans-serif; line-height: 1.6; margin: 0 0 14px;\"\u003eThe \u003cstrong\u003eNVIDIA DGX B200\u003c\/strong\u003e is a unified \u003cstrong\u003eAI platform for training, fine-tuning and inference\u003c\/strong\u003e, powered by \u003cstrong\u003e8x NVIDIA Blackwell B200 GPUs\u003c\/strong\u003e with \u003cstrong\u003e1,440GB of HBM3e\u003c\/strong\u003e GPU memory, \u003cstrong\u003e2x Intel Xeon Platinum 8570\u003c\/strong\u003e processors and \u003cstrong\u003e2TB of system memory\u003c\/strong\u003e, delivering up to \u003cstrong\u003e72 petaFLOPS\u003c\/strong\u003e of FP8 training and \u003cstrong\u003e144 petaFLOPS\u003c\/strong\u003e of FP4 inference performance.\u003c\/p\u003e\n\u003ch2 style=\"font-family: Arial, sans-serif; font-size: 18px; margin: 18px 0 8px;\"\u003eKey Features\u003c\/h2\u003e\n\u003cul style=\"font-family: Arial, sans-serif; line-height: 1.7; margin: 0 0 18px; padding-inline-start: 20px;\"\u003e\n\u003cli\u003e8x NVIDIA Blackwell B200 GPUs with 1,440GB total HBM3e\u003c\/li\u003e\n\u003cli\u003eUp to 72 PFLOPS FP8 training \/ 144 PFLOPS FP4 inference\u003c\/li\u003e\n\u003cli\u003eNVLink 5th generation with NVSwitch for all-to-all GPU bandwidth\u003c\/li\u003e\n\u003cli\u003e2x Intel Xeon Platinum 8570 (112 cores total)\u003c\/li\u003e\n\u003cli\u003e2TB DDR5 system memory (configurable up to 4TB)\u003c\/li\u003e\n\u003cli\u003eNVIDIA ConnectX-7 400Gb\/s and BlueField-3 DPU networking\u003c\/li\u003e\n\u003cli\u003eNVMe storage for OS and AI datasets\u003c\/li\u003e\n\u003cli\u003eIncludes NVIDIA AI Enterprise and Base Command software stack\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch2 style=\"font-family: Arial, sans-serif; font-size: 18px; margin: 18px 0 8px;\"\u003eFull Specifications\u003c\/h2\u003e\n\u003ctable style=\"width: 100%; border-collapse: collapse; font-family: Arial, sans-serif; margin-bottom: 20px;\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003eGeneral\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eBrand\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003eNVIDIA\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eModel\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003eDGX B200\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003ePart Number\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003eDGX-B200\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eType\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003eAI training \u0026amp; inference system\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eForm Factor\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e10U rack\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003eGPU\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eGPUs\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e8x NVIDIA Blackwell B200\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eTotal GPU Memory\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e1,440GB HBM3e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003ePerformance\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e72 PFLOPS FP8 training, 144 PFLOPS FP4 inference\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eInterconnect\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e5th-gen NVLink with NVSwitch\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003eCPU \u0026amp; Memory\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eCPU\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e2x Intel Xeon Platinum 8570, 112 cores total\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eSystem Memory\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e2TB, configurable up to 4TB\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003eStorage\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eOS\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e2x 1.9TB NVMe M.2\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eInternal\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e8x 3.84TB NVMe U.2\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003eNetworking\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eCompute Fabric\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e8x single-port NVIDIA ConnectX-7 VPI, up to 400Gb\/s InfiniBand\/Ethernet\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eStorage\/Management\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e2x dual-port NVIDIA BlueField-3 DPU\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\u003ctd colspan=\"2\" style=\"background: linear-gradient(90deg, rgb(106, 13, 173), rgb(214, 51, 132)); color: rgb(255, 255, 255); padding: 8px; font-weight: bold;\"\u003ePower\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #1a1a2e; color: #fff; font-weight: 600; white-space: nowrap;\"\u003eMax Power\u003c\/td\u003e\n\u003ctd style=\"padding: 6px 8px; border-bottom: 1px solid #ddd; background: #16213e; color: #fff;\"\u003e~14.3kW\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003ch2 style=\"font-family: Arial, sans-serif; font-size: 18px; margin: 18px 0 8px;\"\u003eWho Is It For?\u003c\/h2\u003e\n\u003cul style=\"font-family: Arial, sans-serif; line-height: 1.7; margin: 0 0 18px; padding-inline-start: 20px;\"\u003e\n\u003cli\u003eEnterprise AI and LLM training teams\u003c\/li\u003e\n\u003cli\u003eResearch institutions and universities\u003c\/li\u003e\n\u003cli\u003eData centres building AI factories\u003c\/li\u003e\n\u003cli\u003eGovernment and national AI programmes\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch2 style=\"font-family: Arial, sans-serif; font-size: 18px; margin: 18px 0 8px;\"\u003eFrequently Asked Questions\u003c\/h2\u003e\n\u003cdiv\u003e\n\u003cp style=\"font-family: Arial, sans-serif; line-height: 1.6; margin: 0 0 14px;\"\u003e\u003cstrong\u003eWhat workloads is it designed for?\u003c\/strong\u003e\u003cbr\u003eLarge-scale AI training, fine-tuning and high-throughput inference, including large language models.\u003c\/p\u003e\n\u003cp style=\"font-family: Arial, sans-serif; line-height: 1.6; margin: 0 0 14px;\"\u003e\u003cstrong\u003eHow much GPU memory does it have?\u003c\/strong\u003e\u003cbr\u003e1,440GB of HBM3e across 8 Blackwell B200 GPUs.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cp style=\"font-family: Arial, sans-serif; line-height: 1.6; margin: 0 0 14px;\"\u003eOrder the NVIDIA DGX B200 from \u003cstrong\u003ePC PARK\u003c\/strong\u003e with warranty and fast delivery across Saudi Arabia.\u003c\/p\u003e\n\u003c\/div\u003e","brand":"PC PARK","offers":[{"title":"Default Title","offer_id":67861935947866,"sku":"DGX-B200","price":3281366.6,"currency_code":"SAR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0708\/9029\/1290\/files\/e9e32ec5-8886-4326-a781-44df87fcae7f.png?v=1793124472","url":"https:\/\/pcpark.sa\/en\/products\/nvidia-dgx-b200","provider":"PC PARK","version":"1.0","type":"link"}