{"id":1675,"date":"2025-10-13T11:00:09","date_gmt":"2025-10-13T11:00:09","guid":{"rendered":"https:\/\/aetherix.nl\/?page_id=1675"},"modified":"2025-11-12T09:00:05","modified_gmt":"2025-11-12T09:00:05","slug":"dgx-spark","status":"publish","type":"page","link":"https:\/\/aetherix.nl\/en\/dgx-spark\/","title":{"rendered":"NVIDIA DGX Spark"},"content":{"rendered":"

NVIDIA DGX Spark<\/h1><\/div>

A Grace Blackwell AI supercomputer on your desk.<\/p>\n<\/div>

<\/i> NVIDIA GB10 Superchip
\n<\/i> 128 GB Coherent Integrated System Memory
\n<\/i> NVIDIA ConnectX Network
\n<\/i> NVIDIA AI Software Stack<\/p>\n<\/div>

Show in store<\/span><\/a><\/div><\/div><\/div><\/div><\/div>

NVIDIA DGX Spark<\/h2><\/div>

Powered by the NVIDIA GB10 Grace Blackwell Superchip, the NVIDIA DGX\u2122 Spark delivers 1 petaFLOP of AI performance in an energy-efficient, compact form factor. With the pre-installed NVIDIA AI software stack and 128 GB of memory, developers can locally prototype, refine, and execute the latest generation reasoning models \u2014 like those from DeepSeek, Meta, Google, and others \u2014 featuring up to 200 billion parameters, and seamlessly deploy them to the data center or the cloud.<\/p>\n<\/div><\/div><\/div><\/div><\/div>

NVIDIA GPU, CPU, Network, and AI Software Technologies<\/h2><\/div><\/div><\/div>
<\/i>

NVIDIA GB10 Superchip<\/h3><\/div>

Experience up to 1 petaFLOP of AI performance with FP4 precision, powered by the NVIDIA Grace Blackwell architecture.<\/p>\n<\/div><\/div><\/div>

<\/i>

128 GB Coherent Integrated System Memory<\/h3><\/div>

Run AI development and test workloads with models of up to 200 billion parameters \u2014 right from your desk \u2014 thanks to the large, integrated system memory.<\/p>\n<\/div><\/div><\/div>

<\/i>

NVIDIA ConnectX Network<\/h3><\/div>

The powerful NVIDIA ConnectX networking technology enables two NVIDIA DGX Spark systems to be interconnected, providing high-bandwidth, low-latency communication for distributed AI workloads.<\/p>\n<\/div><\/div><\/div>

<\/i>

NVIDIA AI Software Stack<\/h3><\/div>

Leverage a complete software stack for generative AI workloads \u2014 including tools, frameworks, libraries, and pre-trained models.<\/p>\n<\/div><\/div><\/div><\/div><\/div>

Exploded View<\/h3><\/div>
<\/span><\/div><\/div><\/div><\/div><\/div>

Specifications<\/h2><\/div>\n
\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Architecture<\/strong><\/td>\nNVIDIA Grace Blackwell<\/td>\n<\/tr>\n
GPU<\/strong><\/td>\nNVIDIA Blackwell Architecture<\/td>\n<\/tr>\n
CPU<\/strong><\/td>\n20 core Arm, 10 Cortex-X925 + 10 Cortex-A725 Arm<\/td>\n<\/tr>\n
CUDA Cores<\/strong><\/td>\nNVIDIA Blackwell Generation<\/td>\n<\/tr>\n
Tensor Cores<\/strong><\/td>\n5th Generation<\/td>\n<\/tr>\n
RT Cores<\/strong><\/td>\n4th Generation<\/td>\n<\/tr>\n
Tensor Performance<\/strong><\/td>\n1 PFLOP<\/td>\n<\/tr>\n
System Memory<\/strong><\/td>\n128 GB LPDDR5x, unified system memory<\/td>\n<\/tr>\n
Memory Interface<\/strong><\/td>\n256-bit<\/td>\n<\/tr>\n
Memory Bandwidth<\/strong><\/td>\n273 GB\/s<\/td>\n<\/tr>\n
Storage<\/strong><\/td>\n1 or 4 TB NVME.M2 with self-encryption<\/td>\n<\/tr>\n
USB<\/strong><\/td>\n4x USB TypeC<\/td>\n<\/tr>\n
Ethernet<\/strong><\/td>\n1x RJ-45 connector 10 GbE<\/td>\n<\/tr>\n
NIC<\/strong><\/td>\nConnectX-7 Smart NIC<\/td>\n<\/tr>\n
Wi-Fi<\/strong><\/td>\nWiFi 7<\/td>\n<\/tr>\n
Bluetooth<\/strong><\/td>\nBT 5.3<\/td>\n<\/tr>\n
Audio-output<\/strong><\/td>\nHDMI multichannel audio output<\/td>\n<\/tr>\n
Power Consumption<\/strong><\/td>\nTBD<\/td>\n<\/tr>\n
Display Connectors<\/strong><\/td>\n1x HDMI 2.1a<\/td>\n<\/tr>\n
NVENC | NVDEC<\/strong><\/td>\n1x | 1x<\/td>\n<\/tr>\n
OS<\/strong><\/td>\nNVIDIA DGX\u2122\u00a0OS<\/td>\n<\/tr>\n
System Dimensions<\/strong><\/td>\n150 mm L x 150 mm W x 50.5 mm H<\/td>\n<\/tr>\n
System Weight<\/strong><\/td>\n1.2 kg<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n

View datasheet <\/i><\/a><\/p>\n<\/div><\/div><\/div><\/div><\/div><\/p>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"100-width.php","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-1675","page","type-page","status-publish","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/pages\/1675","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/comments?post=1675"}],"version-history":[{"count":4,"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/pages\/1675\/revisions"}],"predecessor-version":[{"id":1679,"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/pages\/1675\/revisions\/1679"}],"wp:attachment":[{"href":"https:\/\/aetherix.nl\/en\/wp-json\/wp\/v2\/media?parent=1675"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}