NVIDIA DGX Station GB300

Desktop AI Supercomputer for Teams

The NVIDIA DGX Station GB300 brings AI supercomputing to your desk. Designed for researchers, developers, and organizations that need to run large AI models locally without cloud dependency, this system delivers datacenter-class performance in a form factor that fits in any office environment.

384GB Unified Memory (CPU+GPU coherent via NVLink C2C) Up to 2.5 PetaFLOPS FP4 Desktop Tower Workstation

Starting at $94,231

Overview

Why the NVIDIA DGX Station GB300

With 384GB Unified Memory (CPU+GPU coherent via NVLink C2C) and Up to 2.5 PetaFLOPS FP4 performance, the NVIDIA DGX Station GB300 handles workloads that range from AI model training and inference to scientific computing and real-time visualization.

384GB unified coherent memory - no PCIe bottleneck

72-core Grace CPU eliminates CPU bottleneck for data preprocessing

2.5 PFLOPS FP4 in a desktop form factor

100GbE networking for cluster integration

Whisper-quiet for office deployment

Full NVIDIA AI Enterprise software stack included


Specifications

Technical Specifications

Complete hardware specifications for the NVIDIA DGX Station GB300.

GPUNVIDIA Blackwell Ultra GPU
CPUNVIDIA Grace CPU (72 ARM Neoverse V2 Cores)
Memory384GB Unified Memory (CPU+GPU coherent via NVLink C2C)
StorageUp to 16TB NVMe SSD (4x 4TB)
Networking2x 100GbE (ConnectX-7), USB4, Thunderbolt
InterconnectNVLink C2C between Grace CPU and Blackwell GPU
PerformanceUp to 2.5 PetaFLOPS FP4
PowerApproximately 800W
Form FactorDesktop Tower Workstation
Operating SystemNVIDIA DGX OS (Ubuntu-based Linux)
Software StackNVIDIA AI Enterprise, CUDA 13+, cuDNN, TensorRT, NeMo, Triton Inference Server

Use Cases

What You Can Do with the NVIDIA DGX Station GB300

From AI model training to production inference, the NVIDIA DGX Station GB300 handles a wide range of demanding workloads.

  • Multi-user AI development team workstation
  • Training and fine-tuning models up to 400B parameters
  • AI inference at scale for production workloads
  • Scientific computing and molecular dynamics
  • Air-gapped classified AI environments
  • Content creation with generative AI

Petronella Advantage

Why Buy the NVIDIA DGX Station GB300 from Petronella

We do not just sell hardware. We design, deploy, and manage your AI infrastructure with compliance built in from day one. Our entire team is CMMC-RP certified.

Turnkey deployment with pre-loaded AI models for your industry

CMMC/HIPAA/ITAR compliant configuration and documentation

On-site installation with rack or desk placement

Managed AI infrastructure support contract

Staff training on NVIDIA tools and AI workflows

Integration with existing IT infrastructure and Active Directory


Compliance

Compliance-Ready AI Infrastructure

Every NVIDIA DGX Station GB300 deployment from Petronella includes compliance documentation and security hardening for your regulatory requirements. Our CMMC-RP certified team ensures your AI infrastructure meets the standards your industry demands.

CMMC Level 2 HIPAA ITAR NIST 800-171

Petronella Technology Group deploys NVIDIA hardware with full compliance documentation, security hardening, and audit-ready configurations. Whether you operate in defense, healthcare, finance, or government, we ensure your AI systems meet the regulatory frameworks that apply to your organization. Our team holds CMMC-RP, CCNA, CWNE, and DFE certifications.


Related Products

Explore Related NVIDIA Products

Compare the NVIDIA DGX Station GB300 with other NVIDIA solutions to find the right fit for your workloads and budget.


Configure Your NVIDIA DGX Station GB300

Talk to our NVIDIA specialists about the right configuration for your workloads, compliance requirements, and budget. We handle everything from procurement to deployment.