#Japan #Tokyo #AI Supercomputer #NVIDIA DGX #SB C&S

SB C&S Introduces the NVIDIA DGX Spark

In a significant announcement, SB C&S Corporation, a pioneer in bringing new technologies to market, is commencing the distribution of the groundbreaking AI supercomputer, NVIDIA DGX Spark, starting from July 23, 2025. This powerful system is equipped with the NVIDIA GB10 Grace Blackwell Superchip, designed to cater to the burgeoning demands of AI development and large-scale language models (LLMs).

The evolution of generative AI has presented immense opportunities while simultaneously unveiling various operational challenges. Many enterprises struggle as AI development and large-scale data processing are often conducted in disjointed environments, leading to inefficiencies due to the time and resources required for system integration. In addition, the complexity and rising costs associated with managing multiple systems add further hurdles in the AI implementation journey.

The NVIDIA DGX Spark emerges as a solution to these issues, merging NVIDIA’s high-performance AI infrastructure DGX with the distributed processing capabilities of Apache Spark. This innovative AI computer enables users to efficiently manage both AI workloads and data processing within a singular, integrated platform. This integration allows for a streamlined learning and inference process for AI models, significantly enhancing productivity.

One of the standout features of the NVIDIA DGX Spark is its ability to deliver enterprise-level AI development capabilities in a compact personal workspace. Users can build high-precision AI systems within a significantly reduced timeframe, thereby lowering the barriers to entry for AI development and validation projects. The system's compact form factor allows it to be conveniently placed on a desk, eliminating the need for extensive server room facilities. This makes it particularly attractive for researchers facing space constraints, educational institutions, medical facilities, and various sectors such as manufacturing and finance, all of which seek to minimize initial investment costs.

Main Features of NVIDIA DGX Spark

• - Support for large-scale AI models: Utilizing the NVIDIA Grace Blackwell architecture, this system is capable of executing massive models with up to 200 billion parameters locally. It achieves an impressive AI performance of 1,000 AI TOPS, all within a compact package.
• - Compact, portable model: Weighing just 1.2 kilograms, the NVIDIA DGX Spark is lightweight and can easily operate on a standard 100-volt power supply, making installation straightforward.
• - Comprehensive NVIDIA AI software stack: The system comes equipped with all necessary tools for AI deployment and operation, facilitating easy scaling to cloud solutions and NVIDIA’s DGX SuperPOD™.
• - Equipped with NVIDIA ConnectX®: This feature ensures ultra-fast networking optimized for AI, improving data transfer efficiency and allowing for scalability by connecting multiple DGX Spark systems through the ConnectX-7 interface.

Future Directions

Moving forward, SB C&S is committed to addressing the challenges associated with AI implementation and enhancing Japan's competitive edge in IT while promoting widespread adoption of AI technologies across various industries. For more product details, you can visit the official product page at NVIDIA DGX Spark.

In addition, a webinar discussing the capabilities and applications of the NVIDIA DGX Spark is available for interested audiences, which can be accessed here: Webinar Registration.

For any inquiries regarding the product, please reach out at: [email protected].

Important Notices

• - © 2025 NVIDIA Corporation. All rights reserved. NVIDIA, NVIDIA DGX Spark, NVIDIA DGX SuperPOD, and NVIDIA ConnectX are trademarks or registered trademarks in the United States and/or other countries.
• - SoftBank and its logo are registered trademarks or trademarks of SoftBank Group Corp. in Japan and other countries.
• - Other names mentioned in this announcement may be trademarks of their respective companies.