Skip to main content

Nvidia’s supercomputer may bring on a new era of ChatGPT

Nvidia's CEO showing off the company's Grace Hopper computer.
Nvidia

Nvidia has just announced a new supercomputer that may change the future of AI. The DGX GH200, equipped with nearly 500 times more memory than the systems we’re familiar with now, will soon fall into the hands of Google, Meta, and Microsoft.

The goal? Revolutionizing generative AI, recommender systems, and data processing on a scale we’ve never seen before. Are language models like GPT going to benefit, and what will that mean for regular users?

Recommended Videos

Describing Nvidia’s DGX GH200 requires the use of terms most users never have to deal with. “Exaflop,” for example, because the supercomputer provides 1 exaflop of performance and 144 terabytes of shared memory. Nvidia notes that this means nearly 500 times more memory than in a single Nvidia DGX A100 system.

Get your weekly teardown of the tech behind PC gaming
Check your inbox!

Let’s circle back to the 1 exaflop figure and break it down a little. One exaflop equals a quintillion floating-point operations per second (FLOPs). For comparison, Nvidia’s RTX 4090 can hit around 100 teraflops (TFLOPs) when overclocked. A TFLOP equals one trillion floating-point operations per second. The difference is staggering, but of course, the RTX 4090 is not a data center GPU. The DGX GH200, on the other hand, integrates a substantial number of these high-performance GPUs that don’t belong anywhere near a consumer PC.

Nvidia's Grace Hopper superchip.
Nvidia

The computer is powered by Nvidia’s GH200 Grace Hopper superchips. There are 256 of them in total, which, thanks to Nvidia’s NVLink interconnect technology, are all able to work together as a unified system, essentially creating one massive GPU.

The GH200 superchips used here also don’t need a traditional PCIe connection between the CPU and the GPU. Nvidia says that they’re already equipped with an ARM-based Nvidia Grace CP,U as well as an H100 Tensor Core GPU. Nvidia’s got some fancy chip interconnects going on here too, this time using the NVLink-C2C. As a result, the bandwidth between the processor and the graphics card is said to be significantly improved (up to 7 times) and more power-efficient (up to 5 times).

Packing over 200 of these chips into a single powerhouse of a supercomputer is impressive enough, but it gets even better when you consider that, previously, only eight GPUs could be joined with NVLink at a time. A leap from eight to 256 chips certainly gives Nvidia some bragging rights.

It’s hard not to imagine that the DGX GH200 could power improvements in Bard, ChatGPT, and Bing Chat.

Now, where will the DGX GH200 end up and what can it offer to the world? Nvidia’s building its own Helios Supercomputer as a means of advancing its AI research and development. It will encompass four DGX GH200 systems, all interconnected with Nvidia’s Quantum-2 InfiniBand. It expects it to come online by the end of the year.

Nvidia is also sharing its new development with the world, starting with Google Cloud, Meta, and Microsoft. The purpose is much the same — exploring generative AI workloads.

When it comes to Google and Microsoft, it’s hard not to imagine that the DGX GH200 could power improvements in Bard, ChatGPT, and Bing Chat.

Nvidia CEO showing the company's Hopper computer.
Nvidia

The significant computational power provided by a single DGX GH200 system makes it well-suited to advancing the training of sophisticated language models. It’s hard to say what exactly that could mean without comment from one of the interested parties, but we can speculate a little.

More power means larger models, meaning more nuanced and accurate text and a wider range of data for them to be trained on. We might see better cultural understanding, more knowledge of context, and greater coherency. Specialized AI chatbots could also begin popping up, further replacing humans in fields such as technology.

Should we be concerned about potential job displacement, or should we be excited about the advancements these supercomputers could bring? The answer is not straightforward. One thing is for sure — Nvidia’s DGX GH200 might shake things up in the world of AI, and Nvidia has just furthered its AI lead over AMD yet again.

Monica J. White
Monica is a computing writer at Digital Trends, focusing on PC hardware. Since joining the team in 2021, Monica has written…
ChatGPT already listens and speaks. Soon it may see as well
ChatGPT meets a dog

ChatGPT's Advanced Voice Mode, which allows users to converse with the chatbot in real time, could soon gain the gift of sight, according to code discovered in the platform's latest beta build. While OpenAI has not yet confirmed the specific release of the new feature, code in the ChatGPT v1.2024.317 beta build spotted by Android Authority suggests that the so-called "live camera" could be imminently forthcoming.

OpenAI had first shown off Advanced Voice Mode's vision capabilities for ChatGPT in May, when the feature was first launched in alpha. During a demo posted at the time, the system was able to identify that it was looking at a dog through the phone's camera feed, identify the dog based on past interactions, recognize the dog's ball, and associate the dog's relationship to the ball (i.e. playing fetch).

Read more
This open-source alternative to ChatGPT just got serious
The beta Canvas feature on Mistral

French AI startup Mistral announced Monday that it is incorporating a half-dozen new features and capabilities into its free generative AI work assistant, dubbed le Chat (French for "the cat"), that will put the open-source chatbot on par with leading frontier models from OpenAI and Anthropic.

Le Chat can now search the web and provide cited sources, similar to what Perplexity and SearchGPT both offer. Mistral's chatbot now also offers a Canvas feature akin to Claude's Artifacts where users can modify and edit content and code. What's more, le Chat can now generate images thanks to an integration with Black Forest Labs' Flux Pro, the same image generator that powers Grok-2's capabilities.

Read more
This massive upgrade to ChatGPT is coming in January — and it’s not GPT-5
ChatGPT on a laptop

OpenAI is set to launch a new AI agent in January, code-named Operator, that will enable ChatGPT to take action on the user's behalf. You may never have to book your own flights ever again.

The company's leadership made the announcement during a staff meeting Wednesday, reports Bloomberg. The company plans to roll out the new feature as a research preview through the company’s developer API.

Read more