Nvidia (NVDA) on Sunday announced an AI supercomputer called NVIDIA DGX to help develop models for generative AI language applications. Read more here ...

Nvidia has unveiled the DGX GH200 AI supercomputer, which will be capable of up to 1 exaflop of performance ...

Nvidia also said the new Grace Hopper superchip that fuels the DGX GH200 AI supercomputer is in full production mode and systems with the superchip are expected to be available later this year.

But nothing, as far as we can see, that scales down to a single unit 1U or 2U rack server or even something the size of DGX-A100 or DGX-H100 system, which packs two X86 servers and eight GPU ...

Nvidia has just announced the DGX GH200, an AI supercomputer equipped with previously unseen power. It might bring big advancements to generative AI.

Nvidia has seen a ramp-up in orders for its A100 and H100 AI GPUs, as a result of the generative AI boom, which has led to an increase in wafer starts at TSMC, according to market sources.

Clinician-led healthcare artificial intelligence and Nvidia Inception member harisson.ai has deployed eight Nvidia DGX A100 systems in an Equinix International Business Exchange data centre in Sydney.

This makes Sify the first colocation provider in India to offer customers the option to host NVIDIA DGX A100 systems.

The DGX A100 system Nvidia says the requirement could hinder its development of the H100 server accelerator and its ability to support existing A100 customers.

Nvidia claims SPECrate2017_int_base performance of more than 1.5x higher compared to the dualhigh-end AMD Epyc “Rome” generation processors already shipping with Nvidia’s DGX A100 server.