The DGX Spark, formerly known as Project DIGITS, is a compact AI supercomputer designed to provide high-performance computing for AI development. While specific customization options for DGX Spark through partnerships with system builders like ASUS, Dell, HP Inc., and Lenovo are not detailed in the available information, these partnerships generally allow for a range of configurations tailored to meet different user needs. Here are some aspects that might be customizable or influenced by these partnerships:
1. Memory and Storage: The base model of DGX Spark includes 128GB of unified LPDDR5x memory and options for a 1TB or 4TB SSD[7]. System builders might offer additional memory or storage configurations to suit specific requirements.
2. Networking: DGX Spark supports high-performance networking with the NVIDIA ConnectX-7, allowing for the connection of two systems to handle large AI models[9]. Partners might offer different networking configurations or upgrades, such as enhanced connectivity options.
3. Form Factor and Design: While DGX Spark is designed to be compact, system builders could potentially offer custom enclosures or form factors tailored to specific environments or use cases.
4. Software and AI Tools: The system comes with NVIDIA's DGX OS, a custom version of Ubuntu Linux, and pre-installed AI software tools[7]. Partners might offer additional software packages or customizations to support specific AI workflows or applications.
5. Connectivity Options: The system includes features like Wi-Fi 7, USB Type-C, and Bluetooth[7]. System builders could enhance these connectivity options or add others based on user needs.
6. Power Supply and Cooling: Given the high-performance nature of DGX Spark, custom power supply and cooling solutions might be offered by system builders to ensure reliable operation in various environments.
While these customization options are speculative based on typical practices in the industry, the exact details would depend on the specific offerings from each system builder. NVIDIA's partnerships with these builders aim to provide a range of configurations that cater to different users' needs in AI development and research.
Citations:
[1] https://nvidianews.nvidia.com/news/nvidia-announces-dgx-spark-and-dgx-station-personal-ai-computers
[2] https://hothardware.com/news/nvidia-project-digits-renamed-dgx-spark-and-dgx-station
[3] https://docs.cpuc.ca.gov/word_pdf/AGENDA_DECISION/107340.pdf
[4] https://www.engineering.com/nvidia-unveils-dgx-personal-ai-supercomputers-by-grace-blackwell/
[5] https://www.ainvest.com/news/nvidia-unveils-dgx-spark-dgx-station-revolutionizing-personal-ai-computing-2503/
[6] https://www.fibermall.com/blog/dgx-gh200.htm
[7] https://www.pcmag.com/news/what-is-nvidias-dgx-station-a-new-specialized-desktop-line-for-ai-work
[8] https://markets.businessinsider.com/news/stocks/nvidia-announces-dgx-spark-dgx-station-personal-ai-computers-1034490570
[9] https://www.nvidia.com/en-us/products/workstations/dgx-spark/
[10] https://arstechnica.com/ai/2025/03/nvidia-announces-dgx-desktop-personal-ai-supercomputers/
[11] https://www.techpowerup.com/334300/nvidia-announces-dgx-spark-and-dgx-station-personal-ai-computers