DGX-1: World’s First Deep Learning Supercomputer in Box

Share it with your friends Like

Thanks! Share it with your friends!


The NVIDIA® DGX™ is the world's first purpose-built system for deep learning with fully integrated hardware and software that can be deployed quickly and easily. Its revolutionary performance significantly accelerates training time, making the NVIDIA DGX the world's first deep learning supercomputer in a box. Learn more at http://www.nvidia.com/object/deep-learning-system.html.


Steve Chaszar says:

remarkable presentation… thank you for creating

Goku KN says:

What about the coolers for cpu..? How does it work on cooling the cpu..?

crazieeez says:

The huge heatsink, dang! Can't Nvidia make their CPU efficient so that all energy is used for processing instead of dissipating as heat. Seems like their CPU designers aren't smart.


First step on Mars confirmed.

bemused says:

Will i be able to buy a MSI branded "gaming" version through the GPP?

drifter13 says:

How many kidneys does it cost?

kom jojo says:

Mining with this?

Anonymous Eighteen says:

They act like it’s so expensive to make and like it’s so sophisticated just to make money off dirt cheap 10 gb cards costing 1000$ ea

Logan .Kes says:

Im going to mine etherium on it brb…..

Piotr Lenarczyk says:

Google would implement efficient string search on a few of DGX's -despite the fact that it is overpriced approx. 4 times ( please refer to flood with flamable isopropyl or vaseline oil TYAN FT77C-B7079; populated with 8xGTX1080ti ).
I am not a fan of FPGA's, but as far as I do know Neural Networks ( NN's ) are better suited for FPGA's ( polish ARUZ supercomputer in Łódz – if one gently ask, they will help with massive parallelism ).

EpicEnej ! says:

fix geforce shadowplay!!!

xXx_MlgTr1kSh00ter _Xx says:

no shitsherlok, this shit is amazing but nobody review it 🙁

CTElectricGaming says:

cant wait to play pong on it

Akshay Aradhya says:

Btw it costs $129,000

You are better off renting an Amazon Web Server

Osmosis says:

Is deep learning kinda like the human ability to find connection between random images?
Lets say, all images were themed, pirates

Or if im wrong how should i look at "deep learning" ?

Bitcoin says:

28K CUDA this is amazing. I can render VRAY scen in 8K with Photon Maping for 1 sec .AWESOME!!!

EQLuke says:

Maybe nvidia can use this to finally find out the meaning of msrp

PuceN says:

And AMD fans still say nvidia is worse

Write a comment