NVIDIA DGX Station

Personal AI Supercomputer

Groundbreaking AI, at your desk.

Designed for your office, NVIDIA DGX Station is the world’s first personal supercomputer for leading-edge AI development. Built on the same NGC Deep Learning Stack powering all DGX systems, you can now experiment at your desk and extend that same deep learning software across DGX systems and the cloud.

Revolutionary form factor

DGX Station is the only AI supercomputer designed for use at your desk. Whisper quiet at 1/10th the noise of other deep learning workstations, it’s designed for easy experimentation at the office.

Get a Faster Start in Deep Learning

Save up to a month in setup time with DGX Station. DGX Station delivers a rapid plug-in, power-up, deployment experience that’s simple, intuitive, with an integrated hardware and software solution that lets you spend more time focused on insight, instead of piecing together components.

Productivity from Desk to Data Center

Built on the same deep learning software stack and container technology powering all DGX solutions, DGX Station can save you hundreds of thousands of dollars in engineering hours and lost productivity waiting for stable versions of open source frameworks. Use DGX Station to jump immediately into productive experimentation at your desk, and then deploy your work in the data center on DGX-1, or in the cloud.

Supercomputer Performance at Your Desk

DGX Station supports accelerated analytics, deep learning training, and inferencing, all within a single versatile platform, that packs the computing capacity 400 GPU’s into a desk-friendly package that consumes less than 1/20th the power. It’s the only personal AI supercomputer that integrates four Volta powered Tesla V100 GPUs, with innovations including next generation of NVIDIA NVLink and new Tensor Core architecture. DGX Station delivers 3x the training performance of today’s fastest workstations, with 480 TFLOPS of water-cooled performance and FP16 precision.

[please contact us for pricing information]

Partners: