{"product_id":"nvidia-dgx-spark-ai-supercomputer","title":"NVIDIA DGX Spark AI Supercomputer GP 10 Superchip 128GB Memory","description":"\u003cstyle\u003e\n  .sd-dgx-product {\n    max-width: 1220px;\n    margin: 0 auto;\n    padding: 38px 18px 56px;\n    font-family: inherit;\n    color: #161616;\n  }\n\n  .sd-dgx-product * {\n    box-sizing: border-box;\n  }\n\n  .sd-dgx-hero {\n    position: relative;\n    overflow: hidden;\n    border-radius: 34px;\n    background:\n      radial-gradient(circle at 16% 15%, rgba(220, 178, 93, 0.22), transparent 34%),\n      radial-gradient(circle at 84% 42%, rgba(255, 255, 255, 0.10), transparent 30%),\n      linear-gradient(135deg, #302a1e 0%, #111111 45%, #030303 100%);\n    color: #fff;\n    padding: 68px 56px;\n    display: grid;\n    grid-template-columns: 0.96fr 1.04fr;\n    gap: 54px;\n    align-items: center;\n    min-height: 560px;\n    box-shadow: 0 28px 80px rgba(0,0,0,0.12);\n  }\n\n  .sd-dgx-hero::after {\n    content: \"\";\n    position: absolute;\n    inset: 0;\n    background:\n      linear-gradient(90deg, rgba(0,0,0,0.10), rgba(255,255,255,0.025)),\n      linear-gradient(180deg, rgba(255,255,255,0.035), transparent 40%);\n    pointer-events: none;\n  }\n\n  .sd-dgx-hero-copy,\n  .sd-dgx-hero-media {\n    position: relative;\n    z-index: 2;\n  }\n\n  .sd-dgx-eyebrow {\n    display: inline-flex;\n    align-items: center;\n    padding: 8px 16px;\n    border: 1px solid rgba(226, 188, 105, 0.48);\n    border-radius: 999px;\n    color: #e8c672;\n    background: rgba(255,255,255,0.035);\n    font-size: 14px;\n    line-height: 1;\n    letter-spacing: 0.02em;\n    margin-bottom: 26px;\n    backdrop-filter: blur(12px);\n  }\n\n  .sd-dgx-title {\n    color: #ffffff !important;\n    font-size: clamp(42px, 5vw, 72px);\n    line-height: 0.98;\n    letter-spacing: -0.055em;\n    margin: 0 0 24px;\n    font-weight: 760;\n    max-width: 620px;\n  }\n\n  .sd-dgx-title .gold {\n    color: #e9c777;\n  }\n\n  .sd-dgx-subtitle {\n    max-width: 650px;\n    color: rgba(255,255,255,0.74);\n    font-size: 18px;\n    line-height: 1.72;\n    margin: 0;\n  }\n\n  .sd-dgx-stats {\n    display: grid;\n    grid-template-columns: repeat(4, minmax(0, 1fr));\n    gap: 14px;\n    margin-top: 38px;\n  }\n\n  .sd-dgx-stat {\n    min-height: 128px;\n    border: 1px solid rgba(255,255,255,0.13);\n    background: linear-gradient(180deg, rgba(255,255,255,0.085), rgba(255,255,255,0.035));\n    border-radius: 20px;\n    padding: 22px 20px;\n    box-shadow:\n      inset 0 1px 0 rgba(255,255,255,0.08),\n      0 18px 38px rgba(0,0,0,0.10);\n  }\n\n  .sd-dgx-stat strong {\n    display: block;\n    font-size: 28px;\n    line-height: 1;\n    color: #f3d68f;\n    margin-bottom: 13px;\n    letter-spacing: -0.03em;\n  }\n\n  .sd-dgx-stat span {\n    display: block;\n    color: rgba(255,255,255,0.66);\n    font-size: 14px;\n    line-height: 1.38;\n  }\n\n  .sd-dgx-hero-media {\n    display: flex;\n    justify-content: center;\n  }\n\n  .sd-dgx-image-frame {\n    width: 100%;\n    max-width: 570px;\n    border-radius: 30px;\n    padding: 10px;\n    background: linear-gradient(145deg, rgba(255,255,255,0.16), rgba(255,255,255,0.025));\n    border: 1px solid rgba(255,255,255,0.12);\n    box-shadow:\n      0 34px 100px rgba(0,0,0,0.42),\n      0 0 80px rgba(205, 158, 74, 0.12);\n  }\n\n  .sd-dgx-hero-img {\n    width: 100%;\n    display: block;\n    border-radius: 23px;\n    aspect-ratio: 16 \/ 10;\n    object-fit: cover;\n  }\n\n  .sd-dgx-section {\n    margin-top: 34px;\n    border-radius: 30px;\n    background: #f7f4ee;\n    padding: 42px;\n    border: 1px solid #ebe3d7;\n  }\n\n  .sd-dgx-section.white {\n    background: #ffffff;\n  }\n\n  .sd-dgx-section.dark {\n    background:\n      radial-gradient(circle at top right, rgba(223, 180, 92, 0.16), transparent 32%),\n      linear-gradient(135deg, #171717 0%, #080808 100%);\n    color: #ffffff;\n    border-color: #1f1f1f;\n  }\n\n  .sd-dgx-header {\n    max-width: 790px;\n    margin-bottom: 28px;\n  }\n\n  .sd-dgx-kicker {\n    color: #9b7434;\n    font-size: 13px;\n    text-transform: uppercase;\n    letter-spacing: 0.13em;\n    font-weight: 760;\n    margin-bottom: 11px;\n  }\n\n  .sd-dgx-section.dark .sd-dgx-kicker {\n    color: #e6bf6e;\n  }\n\n  .sd-dgx-heading {\n    font-size: clamp(27px, 3vw, 42px);\n    line-height: 1.12;\n    letter-spacing: -0.035em;\n    margin: 0 0 13px;\n    color: inherit;\n  }\n\n  .sd-dgx-text {\n    font-size: 16px;\n    line-height: 1.75;\n    color: #555;\n    margin: 0;\n  }\n\n  .sd-dgx-section.dark .sd-dgx-text {\n    color: rgba(255,255,255,0.72);\n  }\n\n  .sd-dgx-grid-2 {\n    display: grid;\n    grid-template-columns: 1fr 1fr;\n    gap: 30px;\n    align-items: center;\n  }\n\n  .sd-dgx-grid-3 {\n    display: grid;\n    grid-template-columns: repeat(3, 1fr);\n    gap: 18px;\n  }\n\n  .sd-dgx-grid-4 {\n    display: grid;\n    grid-template-columns: repeat(4, 1fr);\n    gap: 16px;\n  }\n\n  .sd-dgx-card {\n    background: #fff;\n    border: 1px solid #ebe3d7;\n    border-radius: 24px;\n    padding: 25px;\n    box-shadow: 0 18px 44px rgba(0,0,0,0.035);\n  }\n\n  .sd-dgx-card.dark-card {\n    background: rgba(255,255,255,0.06);\n    border-color: rgba(255,255,255,0.12);\n    box-shadow: inset 0 1px 0 rgba(255,255,255,0.06);\n  }\n\n  .sd-dgx-card h3 {\n    margin: 0 0 10px;\n    font-size: 19px;\n    line-height: 1.25;\n    letter-spacing: -0.025em;\n    color: inherit;\n  }\n\n  .sd-dgx-card p {\n    margin: 0;\n    color: #5d5d5d;\n    line-height: 1.65;\n    font-size: 15px;\n  }\n\n  .sd-dgx-card.dark-card p {\n    color: rgba(255,255,255,0.68);\n  }\n\n  .sd-dgx-img {\n    width: 100%;\n    display: block;\n    border-radius: 24px;\n  }\n\n  .sd-dgx-feature-img {\n    width: 100%;\n    display: block;\n    border-radius: 18px;\n    margin-bottom: 18px;\n    aspect-ratio: 4 \/ 3;\n    object-fit: cover;\n  }\n\n  .sd-dgx-package-item {\n    padding: 20px 18px;\n    border-radius: 20px;\n    background: #fff;\n    border: 1px solid #ebe3d7;\n    font-weight: 700;\n    text-align: center;\n    box-shadow: 0 14px 34px rgba(0,0,0,0.035);\n  }\n\n  .sd-dgx-note {\n    border-radius: 22px;\n    padding: 22px 24px;\n    background: #fff7e4;\n    border: 1px solid #ead29c;\n    color: #553b10;\n    line-height: 1.65;\n    margin-top: 24px;\n  }\n\n  .sd-dgx-specs {\n    width: 100%;\n    border-collapse: separate;\n    border-spacing: 0;\n    overflow: hidden;\n    border-radius: 24px;\n    border: 1px solid #e8e0d4;\n    background: #fff;\n  }\n\n  .sd-dgx-specs th,\n  .sd-dgx-specs td {\n    padding: 17px 20px;\n    border-bottom: 1px solid #eee7dd;\n    text-align: left;\n    vertical-align: top;\n    font-size: 15px;\n    line-height: 1.58;\n  }\n\n  .sd-dgx-specs th {\n    width: 32%;\n    background: #faf7f1;\n    color: #6e5124;\n    font-weight: 760;\n  }\n\n  .sd-dgx-specs td {\n    color: #151515;\n    background: #ffffff;\n  }\n\n  .sd-dgx-specs tr:last-child th,\n  .sd-dgx-specs tr:last-child td {\n    border-bottom: none;\n  }\n\n  .sd-dgx-links {\n    display: flex;\n    flex-wrap: wrap;\n    gap: 14px;\n    margin-top: 28px;\n    align-items: center;\n  }\n\n  .sd-dgx-product a.sd-dgx-link,\n  .sd-dgx-product a.sd-dgx-link:link,\n  .sd-dgx-product a.sd-dgx-link:visited {\n    display: inline-flex !important;\n    align-items: center !important;\n    justify-content: center !important;\n    min-height: 46px !important;\n    padding: 13px 20px !important;\n    border-radius: 999px !important;\n    background: linear-gradient(135deg, #171717 0%, #2a251b 100%) !important;\n    color: #ffffff !important;\n    border: 1px solid rgba(205, 160, 75, 0.36) !important;\n    text-decoration: none !important;\n    font-size: 14px !important;\n    font-weight: 760 !important;\n    line-height: 1.1 !important;\n    box-shadow: 0 14px 34px rgba(0,0,0,0.14) !important;\n    transition: transform 0.18s ease, box-shadow 0.18s ease, background 0.18s ease !important;\n    white-space: nowrap !important;\n  }\n\n  .sd-dgx-product a.sd-dgx-link:hover {\n    transform: translateY(-1px) !important;\n    background: linear-gradient(135deg, #2a251b 0%, #171717 100%) !important;\n    box-shadow: 0 18px 42px rgba(0,0,0,0.18) !important;\n    color: #f4d990 !important;\n    text-decoration: none !important;\n  }\n\n  .sd-dgx-product a.sd-dgx-link::after,\n  .sd-dgx-product a.sd-dgx-link::before {\n    display: none !important;\n    content: none !important;\n  }\n\n  @media (max-width: 980px) {\n    .sd-dgx-hero,\n    .sd-dgx-grid-2 {\n      grid-template-columns: 1fr;\n    }\n\n    .sd-dgx-hero {\n      padding: 46px 28px;\n      gap: 36px;\n      min-height: auto;\n    }\n\n    .sd-dgx-title {\n      font-size: clamp(38px, 9vw, 58px);\n    }\n\n    .sd-dgx-stats,\n    .sd-dgx-grid-4 {\n      grid-template-columns: repeat(2, 1fr);\n    }\n\n    .sd-dgx-grid-3 {\n      grid-template-columns: 1fr;\n    }\n\n    .sd-dgx-image-frame {\n      max-width: 100%;\n    }\n  }\n\n  @media (max-width: 620px) {\n    .sd-dgx-product {\n      padding: 24px 12px 42px;\n    }\n\n    .sd-dgx-hero {\n      border-radius: 24px;\n      padding: 34px 20px;\n    }\n\n    .sd-dgx-eyebrow {\n      font-size: 12px;\n      padding: 7px 13px;\n      margin-bottom: 20px;\n    }\n\n    .sd-dgx-title {\n      letter-spacing: -0.045em;\n      margin-bottom: 18px;\n    }\n\n    .sd-dgx-subtitle {\n      font-size: 15px;\n      line-height: 1.65;\n    }\n\n    .sd-dgx-stats {\n      grid-template-columns: 1fr 1fr;\n      gap: 10px;\n      margin-top: 28px;\n    }\n\n    .sd-dgx-stat {\n      min-height: 112px;\n      padding: 18px 15px;\n      border-radius: 16px;\n    }\n\n    .sd-dgx-stat strong {\n      font-size: 23px;\n    }\n\n    .sd-dgx-stat span {\n      font-size: 12.5px;\n    }\n\n    .sd-dgx-section {\n      padding: 28px 20px;\n      border-radius: 24px;\n      margin-top: 24px;\n    }\n\n    .sd-dgx-grid-4 {\n      grid-template-columns: 1fr;\n    }\n\n    .sd-dgx-specs th,\n    .sd-dgx-specs td {\n      display: block;\n      width: 100%;\n      padding: 14px 16px;\n    }\n\n    .sd-dgx-specs th {\n      padding-bottom: 5px;\n    }\n\n    .sd-dgx-specs td {\n      padding-top: 5px;\n    }\n\n    .sd-dgx-links {\n      display: grid;\n      grid-template-columns: 1fr;\n      gap: 12px;\n    }\n\n    .sd-dgx-product a.sd-dgx-link {\n      width: 100% !important;\n      white-space: normal !important;\n      text-align: center !important;\n    }\n  }\n\u003c\/style\u003e\n\u003cdiv class=\"sd-dgx-product\"\u003e\n\u003csection class=\"sd-dgx-hero\"\u003e\n\u003cdiv class=\"sd-dgx-hero-copy\"\u003e\n\u003cdiv class=\"sd-dgx-eyebrow\"\u003eIn Stock · Desktop AI Supercomputer\u003c\/div\u003e\n\u003ch1 class=\"sd-dgx-title\"\u003eNVIDIA DGX™ \u003cspan class=\"gold\"\u003eSpark\u003c\/span\u003e\n\u003c\/h1\u003e\n\u003cp class=\"sd-dgx-subtitle\"\u003eA compact AI supercomputer built for local prototyping, fine-tuning, and inference. Powered by the NVIDIA GB10 Grace Blackwell Superchip, DGX™ Spark brings serious AI compute to a refined 1.1L desktop form factor.\u003c\/p\u003e\n\u003cdiv class=\"sd-dgx-stats\"\u003e\n\u003cdiv class=\"sd-dgx-stat\"\u003e\n\u003cstrong\u003e1,000\u003c\/strong\u003e \u003cspan\u003eAI TOPS FP4 Performance\u003c\/span\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-stat\"\u003e\n\u003cstrong\u003e128GB\u003c\/strong\u003e \u003cspan\u003eCoherent Unified Memory\u003c\/span\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-stat\"\u003e\n\u003cstrong\u003e4TB\u003c\/strong\u003e \u003cspan\u003eHigh-Speed NVMe Storage\u003c\/span\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-stat\"\u003e\n\u003cstrong\u003e200B\u003c\/strong\u003e \u003cspan\u003eParameter Models Locally\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-hero-media\"\u003e\n\u003cdiv class=\"sd-dgx-image-frame\"\u003e\u003cimg class=\"sd-dgx-hero-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_5.jpg\" alt=\"PNY NVIDIA DGX Spark AI supercomputer\"\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section white\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003ePackage Contains\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eEverything included to get started\u003c\/h2\u003e\n\u003cp class=\"sd-dgx-text\"\u003eDGX™ Spark includes the core hardware and essential setup materials required for deployment.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-grid-4\"\u003e\n\u003cdiv class=\"sd-dgx-package-item\"\u003eNVIDIA® DGX Spark™\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-package-item\"\u003eQuick Start Card\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-package-item\"\u003eSupport Card\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-package-item\"\u003ePower Supply\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section\"\u003e\n\u003cdiv class=\"sd-dgx-grid-2\"\u003e\n\u003cdiv\u003e\u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_7a.jpg\" alt=\"DGX Spark AI workstation\"\u003e\u003c\/div\u003e\n\u003cdiv\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eOverview\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eBuilt for serious AI development on your desk\u003c\/h2\u003e\n\u003cp class=\"sd-dgx-text\"\u003eNVIDIA DGX™ Spark is designed for developers, researchers, startups, enterprises, and AI teams who need to prototype, fine-tune, and run large AI models locally. With the NVIDIA AI software stack preinstalled, it offers a streamlined path from desktop experimentation to deployment in the data center or cloud.\u003c\/p\u003e\n\u003cdiv class=\"sd-dgx-note\"\u003e\n\u003cstrong\u003eImportant:\u003c\/strong\u003e DGX™ Spark is not a Windows® PC. It is designed as an AI supercomputer with NVIDIA DGX™ OS, Ubuntu Linux, and the NVIDIA AI software stack.\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section dark\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eKey Features\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eCompact form. Enterprise-class AI architecture.\u003c\/h2\u003e\n\u003cp class=\"sd-dgx-text\"\u003eBuilt around NVIDIA Grace Blackwell, DGX™ Spark delivers the memory, compute, networking, and software foundation required for modern generative AI development.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-grid-3\"\u003e\n\u003cdiv class=\"sd-dgx-card dark-card\"\u003e\n\u003cimg class=\"sd-dgx-feature-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_17a.jpg\" alt=\"NVIDIA GB10 Superchip\"\u003e\n\u003ch3\u003eNVIDIA GB10 Superchip\u003c\/h3\u003e\n\u003cp\u003eUp to 1,000 AI TOPS of FP4 performance, powered by NVIDIA Grace Blackwell architecture.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-card dark-card\"\u003e\n\u003cimg class=\"sd-dgx-feature-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_18a.jpg\" alt=\"128GB unified memory\"\u003e\n\u003ch3\u003e128GB Unified Memory\u003c\/h3\u003e\n\u003cp\u003eCoherent system memory enables large AI workloads and local model development up to 200B parameters.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-card dark-card\"\u003e\n\u003cimg class=\"sd-dgx-feature-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_19a.jpg\" alt=\"NVIDIA ConnectX networking\"\u003e\n\u003ch3\u003eNVIDIA ConnectX Networking\u003c\/h3\u003e\n\u003cp\u003eConnect two DGX™ Spark systems together to expand supported model capacity up to 405B parameters.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section white\"\u003e\n\u003cdiv class=\"sd-dgx-grid-2\"\u003e\n\u003cdiv\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eAI Workflow\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eDevelop locally. Deploy anywhere.\u003c\/h2\u003e\n\u003cp class=\"sd-dgx-text\"\u003eDGX™ Spark gives AI teams a powerful local environment for prototyping, fine-tuning, inference, and experimentation. It mirrors the NVIDIA software architecture used across larger AI infrastructure, making it easier to move work from desktop development to cloud, data center, or accelerated production environments.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv\u003e\u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_8.jpg\" alt=\"DGX Spark product detail\"\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003ePerformance Foundation\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eDesigned for large-parameter AI models\u003c\/h2\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-grid-3\"\u003e\n\u003cdiv class=\"sd-dgx-card\"\u003e\n\u003ch3\u003eGrace Blackwell Architecture\u003c\/h3\u003e\n\u003cp\u003eThe GB10 Superchip combines NVIDIA Blackwell GPU architecture with a high-performance 20-core Arm CPU for AI model orchestration, preprocessing, and inference workloads.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-card\"\u003e\n\u003ch3\u003e200B Models Locally\u003c\/h3\u003e\n\u003cp\u003eWith 128GB of coherent unified memory and FP4 support, DGX™ Spark can run AI models up to 200 billion parameters directly on the desktop.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-card\"\u003e\n\u003ch3\u003eScale to 405B Parameters\u003c\/h3\u003e\n\u003cp\u003eTwo DGX™ Spark systems can be linked using NVIDIA ConnectX networking to support larger inference workloads up to 405 billion parameters.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section white\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eGallery\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eRefined hardware. Serious compute density.\u003c\/h2\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-grid-2\"\u003e\n\u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_9.jpg\" alt=\"DGX Spark angle view\"\u003e \u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_10.jpg\" alt=\"DGX Spark product view\"\u003e \u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_11.jpg\" alt=\"DGX Spark rear view\"\u003e \u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_12.jpg\" alt=\"DGX Spark detail view\"\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eI\/O Ports\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003eHigh-speed connectivity for AI workflows\u003c\/h2\u003e\n\u003cp class=\"sd-dgx-text\"\u003eDesigned with high-bandwidth networking, USB 4 connectivity, HDMI output, and a compact desktop footprint.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cimg class=\"sd-dgx-img\" src=\"https:\/\/media.memoryexpress.com\/Assets\/Products\/MX00134639\/PNY-DGX-Spark_io.jpg\" alt=\"PNY NVIDIA DGX Spark I\/O ports\"\u003e\u003c\/section\u003e\n\u003csection class=\"sd-dgx-section white\"\u003e\n\u003cdiv class=\"sd-dgx-header\"\u003e\n\u003cdiv class=\"sd-dgx-kicker\"\u003eSpecifications\u003c\/div\u003e\n\u003ch2 class=\"sd-dgx-heading\"\u003ePNY NVIDIA DGX™ Spark technical details\u003c\/h2\u003e\n\u003c\/div\u003e\n\u003ctable class=\"sd-dgx-specs\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eMake and Model\u003c\/th\u003e\n\u003ctd\u003ePNY NVIDIA DGX™ Spark\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePart Number\u003c\/th\u003e\n\u003ctd\u003eNVDGXSPARK-PB\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eColour\u003c\/th\u003e\n\u003ctd\u003eGold\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eOperating System\u003c\/th\u003e\n\u003ctd\u003eNVIDIA DGX™ OS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eCPU \/ GPU\u003c\/th\u003e\n\u003ctd\u003eNVIDIA GB10 Superchip\u003cbr\u003eCPU: 20-core Arm processor with 10× Cortex-X925 and 10× Cortex-A725 cores\u003cbr\u003eGPU: NVIDIA Blackwell architecture with 5th Gen Tensor Cores and 4th Gen RT Cores\u003cbr\u003eTensor Performance: Up to 1,000 AI TOPS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eMemory\u003c\/th\u003e\n\u003ctd\u003e128GB LPDDR5x 256-bit coherent unified system memory\u003cbr\u003eMemory Bandwidth: 273GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eStorage\u003c\/th\u003e\n\u003ctd\u003e4TB NVMe M.2 with self-encryption\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eAudio\u003c\/th\u003e\n\u003ctd\u003eAudio out via HDMI port only\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eVideo Encode \/ Decode\u003c\/th\u003e\n\u003ctd\u003e1× NVENC · 1× NVDEC\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eWired Networking\u003c\/th\u003e\n\u003ctd\u003eNVIDIA ConnectX-7 LAN chipset\u003cbr\u003e1× RJ45 10Gb Ethernet LAN port\u003cbr\u003e1× 200Gb NVIDIA ConnectX-7 NIC port\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eWireless\u003c\/th\u003e\n\u003ctd\u003eTri-band IEEE 802.11be Wi-Fi 7\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eBluetooth\u003c\/th\u003e\n\u003ctd\u003eBluetooth v5.3 with LE protocol\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eI\/O Ports\u003c\/th\u003e\n\u003ctd\u003e1× HDMI 2.1a port\u003cbr\u003e1× 10Gb RJ45 Ethernet LAN port\u003cbr\u003e1× 200Gb NVIDIA ConnectX-7 NIC port\u003cbr\u003e4× USB 4 Type-C ports, up to 40Gbps\u003cbr\u003e1× USB 4 Type-C port used for system power\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePower\u003c\/th\u003e\n\u003ctd\u003ePower input via USB 4 Type-C port\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eDimensions\u003c\/th\u003e\n\u003ctd\u003e150 × 150 × 50.5 mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003cdiv class=\"sd-dgx-links\"\u003e\n\u003ca class=\"sd-dgx-link\" href=\"https:\/\/www.pny.com\/professional\/software-solutions\/desktop-ai-supercomputer\" rel=\"noopener\" target=\"_blank\"\u003e Manufacturer Product Page \u003c\/a\u003e \u003ca class=\"sd-dgx-link\" href=\"https:\/\/www.pny.com\/file%20library\/company\/support\/product%20brochures\/nvidia-dgx-spark\/nvidia-dgx-spark-workstation-datasheet.pdf\" rel=\"noopener\" target=\"_blank\"\u003e Download Data Sheet \u003c\/a\u003e \u003ca class=\"sd-dgx-link\" href=\"https:\/\/docs.nvidia.com\/dgx\/dgx-os-6-user-guide\/introduction.html\" rel=\"noopener\" target=\"_blank\"\u003e Learn About DGX™ OS\u003c\/a\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"sd-dgx-links\"\u003e\n\u003ctable border=\"0\" cellpadding=\"0\" cellspacing=\"0\" class=\"pi-specs-table w-100 hidden-md-down\" style=\"width: 99.9327%; height: 20.9271px;\"\u003e\n\u003ctbody\u003e\n\u003ctr class=\"row w-100 mx-0 align-items-center\" id=\"warranty_return_period_1\" style=\"height: 20.9271px;\"\u003e\n\u003ctd class=\"col-6\" style=\"width: 78.7669%; height: 20.9271px;\"\u003e\u003cstrong\u003eManufacturer Warranty\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd class=\"col-6\" style=\"width: 21.0311%; height: 20.9271px;\"\u003e\u003cstrong\u003e1 Year\u003c\/strong\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003cp\u003e\u003cstrong\u003eFinal sale, no refund or exchange\u003c\/strong\u003e\u003c\/p\u003e\n\u003c\/div\u003e\n\u003c\/section\u003e\n\u003c\/div\u003e","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":44915893698697,"sku":"298701","price":7099.0,"currency_code":"CAD","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0665\/1242\/1001\/files\/Screenshot_2026-05-25_011936.png?v=1779686420","url":"https:\/\/www.speedydrone.ca\/products\/nvidia-dgx-spark-ai-supercomputer","provider":"SpeedyDrone Canada ","version":"1.0","type":"link"}