NVIDIA DGX Platform

NVIDIA DGXSpark

Your Personal AI Supercomputer

One PetaFLOP of AI performance on your desk. Run models with up to 200 billion parameters, fine-tune on proprietary data, and develop AI applications -- all without sending a single byte to the cloud.

NVIDIA DGX Spark personal AI supercomputer

1

PetaFLOP FP4

128

GB Unified Memory

200B

Parameter Models

4TB

NVMe Storage

NVIDIA Grace CPUBlackwell GPUWiFi 710GbE
NVIDIA DGX Spark desktop AI computer beauty shot

What You Can Do With DGX Spark

A full AI development platform that fits on your desk. No cloud bills, no data exposure, no GPU rental queues.

Run LLMs Locally

Run large language models with up to 200 billion parameters on your desk. Llama, Mistral, Gemma, and thousands of open-source models at full speed.

Fine-Tune Models

Customize foundation models on your proprietary data. Train domain-specific AI that understands your industry and workflows.

Eliminate Cloud Costs

Stop paying per-token API fees and hourly GPU rentals. DGX Spark is a one-time investment that pays for itself in months.

Private & Secure AI

Your data never leaves your facility. Perfect for classified, regulated, or sensitive workloads where cloud AI is not an option.

Scale with NVLink

Connect two DGX Spark units via NVLink to double memory to 256GB and boost compute. Scale your personal AI lab as needs grow.

AI Development

Build RAG applications, AI agents, computer vision systems, and production AI services. Full CUDA development environment.

Who DGX Spark Is Built For

From individual researchers to compliance-driven organizations, DGX Spark puts enterprise AI in reach.

AI Researchers

Experiment with large models, run ablation studies, and prototype new architectures without cloud GPU queues.

Data Scientists

Develop and test ML pipelines, run feature engineering at scale, and iterate on models faster.

Small AI Teams

A shared AI development platform for prototyping, testing, and small-scale training.

Defense Contractors

Air-gapped AI for classified environments. CMMC-compliant deployment available from PTG.

Healthcare Organizations

HIPAA-compliant AI for patient data analysis, medical imaging, and clinical NLP. All on-premises.

AI Developers

Build production AI applications -- RAG systems, chatbots, code assistants, image generators -- locally.

Technical Specifications

Every detail of the NVIDIA DGX Spark personal AI supercomputer.

AI Performance1 PetaFLOP FP4
GPUNVIDIA Blackwell GPU
CPUNVIDIA Grace CPU20 ARM Cores: 10x Cortex-X925 + 10x Cortex-A725
Memory128GB Unified LPDDR5X273 GB/s bandwidth | Shared between CPU and GPU
StorageUp to 4TB NVMe SSD
Networking1x 10GbE Ethernet
WirelessWiFi 7 + Bluetooth 5.3
Multi-UnitConnect 2 units via NVLinkDouble memory to 256GB and increase compute
Form FactorDesktop AI ComputerCompact, office-friendly design
PowerStandard wall outletNo special electrical requirements
CoolingFan-cooled, office-quiet
SoftwareNVIDIA AI Enterprise, NGC Catalog, NIMCUDA, cuDNN, PyTorch, TensorFlow included

The PTG Advantage

DGX Spark is powerful out of the box. With PTG, it becomes a complete AI solution.

Pre-Configured AI Stack

We set up your preferred frameworks, models, development tools, and workflows before delivery.

CMMC/HIPAA Compliant

Our CMMC-RP certified team configures encryption, access controls, audit logging, and security hardening.

AI Consulting Included

Bundled AI consulting hours. We help you identify use cases, select models, and build your first applications.

On-Site Setup & Training

Raleigh-Durham Triangle: in-person delivery, install, and training. Remote setup available nationwide.

DGX Spark vs. Cloud AI

NVIDIA DGX Spark

One-time purchase, unlimited usage
Data never leaves your facility
Zero latency, always available
CMMC/HIPAA compliant by design
Run any model -- no vendor lock-in

Cloud AI Services

Recurring monthly costs that scale
Data sent to third-party servers
Network latency, outages, rate limits
Compliance complexity with external processors
Limited to provider's model selection

Frequently Asked Questions

What AI models can DGX Spark run?
With 128GB of unified memory, DGX Spark can run LLMs up to 200 billion parameters locally. This includes Llama 3.1 70B, Mixtral 8x22B, and many other open-source and commercial models. You can run inference, fine-tuning, and RAG applications on your desktop.
How does it compare to a GPU workstation?
DGX Spark uses unified memory architecture (CPU and GPU share the same 128GB pool), eliminating the PCIe bottleneck that limits traditional GPU workstations. It delivers 1 PetaFLOP FP4 in a compact desktop using a fraction of the power. See our AI workstations for traditional GPU configurations.
Can I connect multiple DGX Sparks together?
Yes. Two DGX Spark units connect via NVLink to double memory to 256GB and increase compute. PTG can configure and optimize multi-unit setups for your use case.
What software is included?
NVIDIA AI Enterprise, NGC catalog, NIM microservices. PTG additionally pre-configures PyTorch, TensorFlow, Ollama, and your preferred development environments.
Does it need special power or cooling?
No. Standard wall outlet, quiet fan cooling. Designed to sit on or under your desk with no special electrical or HVAC requirements.
Can PTG help with HIPAA/CMMC compliance?
Absolutely. Our CMMC-RP certified team configures encryption, access controls, audit logging, and security hardening for HIPAA, CMMC, and NIST 800-171.
When is DGX Spark available?
Available for pre-order through PTG. Contact us to reserve your unit. We offer bundled packages including setup, training, and AI consulting.

Bring AI to Your Desk

NVIDIA DGX Spark puts a PetaFLOP of AI performance in a desktop form factor. No data center required, no cloud dependency, no compromise on capability.

Reserve your DGX Spark today and our CMMC-RP certified team will configure it for your exact use case.