{"product_id":"nvidia-dgx-spark-ai-supercomputer","title":"NVidia DGX Spark - Personal AI Supercomputer 4TB","description":"\u003ch2 style=\"\n    color: black;\n\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003eNVIDIA DGX Spark - \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eDGX personal AI computer, designed to build \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eand run AI.\u003c\/span\u003e\n\u003c\/h2\u003e\n\u003cp style=\"\n    color: black;\n\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eDevelop Locally, Deploy Anywhere at Scale\u003c\/span\u003e\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003e \u003c\/p\u003e\n\u003cdiv data-loaded=\"true\" aria-label=\"Page 2\" role=\"region\" data-page-number=\"2\" class=\"page\"\u003e\n\u003cdiv class=\"textLayer\" style=\"\n    color: black;\n\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003cstrong\u003eNVIDIA DGX Spark\u003c\/strong\u003e provides \u003cstrong\u003eorganizations\u003c\/strong\u003e and \u003cstrong\u003edevelopers\u003c\/strong\u003e with a powerful, \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eeconomical experimentation ground for prototype models, freeing up valuable \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003ecompute resources in their cluster environments better suited for training \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eand deploying production models. Leveraging the NVIDIA AI platform software \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003earchitecture makes it possible for NVIDIA DGX Spark users to easily move their \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003ework from their desktop to DGX Cloud or any accelerated cloud or data center \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003einfrastructure, making it easier than ever to prototype, fine-tune, and iterate.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"textLayer\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003c\/span\u003e\u003cbr\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"textLayer\" style=\"\n    color: black;\n\"\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003cstrong\u003eDesktop AI Compute Demands\u003c\/strong\u003e\u003cbr role=\"presentation\"\u003eThe increasing size and complexity of generative AI models is making development efforts on local systems challenging. Prototyping, tuning, and inferencing large models locally requires large amounts of memory and significant compute performance. As enterprises, software providers, government agencies, startups, and researchers staff up AI efforts, the need for AI compute resources continues to grow.\u003c\/span\u003e\u003c\/div\u003e\n\u003cdiv class=\"textLayer\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003c\/span\u003e\u003cbr\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"textLayer\" style=\"\n    color: black;\n\"\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003cstrong\u003eWork With Large-Parameter AI Models\u003c\/strong\u003e\u003cbr role=\"presentation\"\u003eWith 128 GB of unified system memory and support for the FP4 data format, NVIDIA DGX Spark can support AI models of up to 200B parameters, enabling AI developers to prototype, fine-tune, and inference large models on their desktop. With built-in NVIDIA ConnectX network technology, two NVIDIA DGX Spark systems can be connected to work on even larger models such as Llama 3.1 405B.\u003c\/span\u003e\u003c\/div\u003e\n\u003cdiv class=\"textLayer\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003e\u003c\/span\u003e\u003cbr\u003e\n\u003c\/div\u003e\n\u003ch3 class=\"textLayer\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eTechnical Specifications:\u003c\/span\u003e\u003c\/strong\u003e\u003c\/h3\u003e\n\u003cdiv class=\"textLayer\"\u003e\n\u003ctable width=\"100%\" style=\"width: 100.036%;\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eArchitecture\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e \u003cspan\u003eNVIDIA Grace Blackwell\u003c\/span\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eGPU\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eNVIDIA Blackwell Architecture\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eCPU\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\n\u003cspan role=\"presentation\" dir=\"ltr\"\u003e20 core Arm, 10 Cortex-X925\u003c\/span\u003e\u003cbr role=\"presentation\"\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003e+ 10 Cortex-A725 Arm\u003c\/span\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eCUDA Cores\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eNVIDIA Blackwell Generation\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eTensor Cores\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e5th Generation\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eRT Cores\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e4th Generation\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eTensor Performance\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e1 PFLOP\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eSystem Memory\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\n\u003cspan role=\"presentation\" dir=\"ltr\"\u003e128 GB LPDDR5x, coherent unified \u003c\/span\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003esystem memory\u003c\/span\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan role=\"presentation\" dir=\"ltr\"\u003eMemory Interface\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e256-bit\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eMemory Bandwidth\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eUp to 273 GB\/s\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eStorage\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e4 TB NVME.M2 with self-encryption\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eUSB\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e4x USB TypeC\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eEthernet\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003e1x RJ-45 connector \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003e10 GbE\u003c\/span\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eNIC\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eConnectX-7 NIC @ 200 Gbps\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eWi-Fi\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eWiFi 7\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eBluetooth\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eBT 5.4 w\/LE\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eAudio-output\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003eHDMI multichannel audio output\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003ePower Supply\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e240 W\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eGB10 TDP*\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e140 W\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eDisplay Connectors\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e1x HDMI 2.1a\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eNVENC | NVDEC\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e1x | 1x\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003eOS\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\n\u003cspan dir=\"ltr\" role=\"presentation\"\u003eNVIDIA DGX\u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003e™\u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003e \u003c\/span\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eOS\u003c\/span\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eSystem Dimensions\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e150 mm L x 150 mm W x 50.5 mm H\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd style=\"width: 32.554%;\"\u003e\u003cstrong\u003e\u003cspan dir=\"ltr\" role=\"presentation\"\u003eSystem Weight\u003c\/span\u003e\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 67.2662%;\"\u003e\u003cspan\u003e1.2 kg\u003c\/span\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv data-loaded=\"true\" aria-label=\"Page 3\" role=\"region\" data-page-number=\"3\" class=\"page\"\u003e\n\u003cdiv class=\"textLayer\"\u003e\u003c\/div\u003e\n\u003c\/div\u003e","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":43063068393520,"sku":"1097764","price":524290.0,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0646\/8855\/4032\/files\/DGX-Spark.png?v=1779456435","url":"https:\/\/mehtabrothers.in\/products\/nvidia-dgx-spark-ai-supercomputer","provider":"Mehta Brothers Shop","version":"1.0","type":"link"}