NVIDIA DGX Spark AI Supercomputer Bundle (2x Systems)

Dual AI Power for Advanced Workloads

The NVIDIA DGX Spark Bundle combines two high-performance DGX Spark systems with a high-speed interconnect cable, delivering a scalable AI computing solution designed for enterprises, developers, and research teams. This dual-system configuration unlocks significantly higher performance, enabling users to train, fine-tune, and deploy large-scale AI models faster and more efficiently.

Built on the advanced Grace Blackwell architecture, each unit features a powerful combination of CPU and GPU integrated into a unified system. When paired together, the bundle provides enhanced compute capabilities, improved memory capacity, and the ability to handle even more complex AI workloads.

Key Features Of NVIDIA DGX Spark

Dual-System AI Performance

• Includes 2× DGX Spark AI supercomputers + high-speed QSFP cable
• Up to 2 PetaFLOPS combined AI performance (FP4)
• Ideal for distributed AI workloads and clustering

Massive Unified Memory for Large Models

• Total 256GB LPDDR5x unified memory (128GB per unit)
• Run and scale large AI models efficiently
• Supports models up to 400B+ parameters when clustered

Grace Blackwell Superchip Architecture

• Each system powered by 20-core Arm CPU + Blackwell GPU
• 5th Gen Tensor Cores for next-gen AI acceleration
• Seamless CPU-GPU integration for maximum efficiency

Enterprise Storage & Speed

• 2× 4TB NVMe SSD (total 8TB storage)
• High-speed access for datasets, models, and pipelines
• Secure self-encrypting storage for enterprise environments

Ultra-Fast Networking & Scaling

• Includes QSFP high-speed interconnect cable (200Gbps)
• ConnectX-7 Smart NIC for clustering and scaling
• Ideal for distributed AI training and inference
  

  Built for Enterprise AI & Innovation

  The DGX Spark Bundle is designed for organizations that require more than a single workstation. By combining two systems, users can:

  • Train larger AI models faster
  • Scale workloads across multiple nodes
  • Run advanced generative AI applications
  • Build enterprise-grade AI infrastructure locally

  With its pre-installed NVIDIA AI software stack, including CUDA, TensorRT, and AI development frameworks, the system enables rapid deployment and reduces setup time significantly.

  Workflows, Models & AI Tools

  The NVIDIA DGX Spark is built to accelerate AI development from day one, providing a complete ecosystem of software, tools, and frameworks required to run advanced workloads seamlessly. Whether you’re building machine learning models, deploying AI applications, or experimenting with generative AI, DGX Spark offers a ready-to-use environment designed for speed and efficiency.

  

  Developers have the flexibility to install and run their preferred frameworks and toolchains, while also benefiting from NVIDIA’s optimized AI ecosystem. This includes access to advanced solutions like NVIDIA NIM, which simplifies model deployment, as well as AI Blueprints for rapid solution development and AI Workbench for streamlined experimentation and workflow management.

  With pre-integrated support for modern AI pipelines, DGX Spark enables users to:

  • Quickly build, train, and fine-tune AI models
  • Deploy scalable AI microservices with minimal setup
  • Optimize workflows using NVIDIA’s performance-tuned tools
  • Transition smoothly from development to production

  This powerful combination of hardware + software integration ensures that developers and enterprises can focus on innovation rather than infrastructure setup, making DGX Spark a complete platform for next-generation AI development.

  Ideal Use Cases NVIDIA DGX Spark

  • Large Language Model (LLM) Training
  • Generative AI Development
  • Enterprise AI Deployment
  • Data Science & Analytics Pipelines
  • Autonomous Systems & Robotics AI