NVIDIA DGX Spark Personal AI Supercomputer

Compact AI Power for Developers & Enterprises

The NVIDIA DGX Spark Personal AI Supercomputer is a cutting-edge desktop AI system designed to bring datacenter-level AI performance directly to your workspace. Built on the advanced GB10 Grace Blackwell Superchip, this compact yet powerful machine enables developers, data scientists, and enterprises to build, train, and deploy AI models locally with unmatched efficiency.

With up to 1 PetaFLOP of AI performance, DGX Spark delivers exceptional compute capabilities for modern workloads such as generative AI, machine learning, and deep learning applications.

Key Features & Highlights

High-Performance AI Computing

✅ Up to 1 PFLOP (FP4) AI performance
✅ Powered by NVIDIA Blackwell GPU with 5th Gen Tensor Cores
✅ Designed for training, inference, and fine-tuning AI models

DGX Spark provides the performance needed to handle complex AI tasks while maintaining a compact desktop footprint.

128GB Unified Memory Architecture

✅ 128GB LPDDR5x unified system memory
✅ Shared between CPU and GPU for seamless data flow
✅Supports AI models up to 200 billion parameters

This unified architecture eliminates traditional bottlenecks, allowing faster processing of large datasets and AI workloads.

GB10 Grace Blackwell Superchip

✅ Integrated 20-core Arm CPU + Blackwell GPU
✅ High-speed interconnect for efficient compute
✅ Optimized for AI workloads and parallel processing

This advanced architecture ensures high efficiency, low latency, and maximum performance for AI development.

Fast & Secure Storage

✅ 4TB NVMe SSD with self-encryption
✅ High-speed access to datasets and models
✅ Ideal for AI pipelines and large-scale data processing

The system ensures secure and reliable storage for enterprise-grade workloads.

Advanced Networking & Connectivity

✅ NVIDIA ConnectX-7 Smart NIC
✅ High-speed networking for scaling and clustering
✅ Supports multi-system connectivity

DGX Spark can be connected with additional systems to scale performance for larger AI workloads.

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.

Built for AI Development & Innovation

The DGX Spark is more than just hardware—it is a complete AI platform equipped with NVIDIA’s full software ecosystem. This allows users to:

• Develop and test AI models locally
• Fine-tune models up to 70 billion parameters
• Run inference on models up to 200 billion parameters
• Build AI agents and real-world applications

Its pre-configured environment accelerates development while reducing setup complexity.