Posts in category: nVidia AI

To understand the latest advance in generative AI, imagine a courtroom.

Judges hear and decide cases based on their general understanding of the law. Sometimes a case — like a malpractice suit or a labor dispute —  requires special expertise, so judges send court clerks to a law library, looking for precedents and specific cases they can cite.

Like a good judge, large language models (LLMs) can respond to a wide variety of human queries. But to deliver authoritative answers that cite sources, the model needs an assistant to do some research.

The court clerk of AI is a process called retrieval-augmented generation, or RAG for short.

The Story of the Name

Patrick Lewis, lead author of the 2020 paper that coined the term, apologized for the unflattering acronym that now describes a growing family of methods across hundreds of papers and dozens of commercial services he believes represent the future of generative AI.

“We definitely would have put more thought into the name had we known our work would become so widespread,” Lewis said in an interview from Singapore, where he was sharing his ideas with a regional conference of database developers.

“We always planned to have a nicer sounding name, but when it came time to write the paper, no one had a better idea,” said Lewis, who now leads a RAG team at AI startup Cohere.

So, What Is Retrieval-Augmented Generation?

Retrieval-augmented generation is a technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources.

In other words, it fills a gap in how LLMs work. Under the hood, LLMs are neural networks, typically measured by how many parameters they contain. An LLM’s parameters essentially represent the general patterns of how humans use words to form sentences.

That deep understanding, sometimes called parameterized knowledge, makes LLMs useful in responding to general prompts at light speed. However, it does not serve users who want a deeper dive into a current or more specific topic.

Combining Internal, External Resources

Lewis and colleagues developed retrieval-augmented generation to link generative AI services to external resources, especially ones rich in the latest technical details.

The paper, with coauthors from the former Facebook AI Research (now Meta AI), University College London and New York University, called RAG “a general-purpose fine-tuning recipe” because it can be used by nearly any LLM to connect with practically any external resource.

Building User Trust

Retrieval-augmented generation gives models sources they can cite, like footnotes in a research paper, so users can check any claims. That builds trust.

What’s more, the technique can help models clear up ambiguity in a user query. It also reduces the possibility a model will make a wrong guess, a phenomenon sometimes called hallucination.

Another great advantage of RAG is it’s relatively easy. A blog by Lewis and three of the paper’s coauthors said developers can implement the process with as few as five lines of code.

That makes the method faster and less expensive than retraining a model with additional datasets. And it lets users hot-swap new sources on the fly.

How People Are Using Retrieval-Augmented Generation 

With retrieval-augmented generation, users can essentially have conversations with data repositories, opening up new kinds of experiences. This means the applications for RAG could be multiple times the number of available datasets.

For example, a generative AI model supplemented with a medical index could be a great assistant for a doctor or nurse. Financial analysts would benefit from an assistant linked to market data.

In fact, almost any business can turn its technical or policy manuals, videos or logs into resources called knowledge bases that can enhance LLMs. These sources can enable use cases such as customer or field support, employee training and developer productivity.

The broad potential is why companies including AWS, IBM, Glean, Google, Microsoft, NVIDIA, Oracle and Pinecone are adopting RAG.

Getting Started With Retrieval-Augmented Generation 

To help users get started, NVIDIA developed a reference architecture for retrieval-augmented generation. It includes a sample chatbot and the elements users need to create their own applications with this new method.

The workflow uses NVIDIA NeMo, a framework for developing and customizing generative AI models, as well as software like NVIDIA Triton Inference Server and NVIDIA TensorRT-LLM for running generative AI models in production.

The software components are all part of NVIDIA AI Enterprise, a software platform that accelerates development and deployment of production-ready AI with the security, support and stability businesses need.

Getting the best performance for RAG workflows requires massive amounts of memory and compute to move and process data. The NVIDIA GH200 Grace Hopper Superchip, with its 288GB of fast HBM3e memory and 8 petaflops of compute, is ideal — it can deliver a 150x speedup over using a CPU.

Once companies get familiar with RAG, they can combine a variety of off-the-shelf or custom LLMs with internal or external knowledge bases to create a wide range of assistants that help their employees and customers.

RAG doesn’t require a data center. LLMs are debuting on Windows PCs, thanks to NVIDIA software that enables all sorts of applications users can access even on their laptops.

PCs equipped with NVIDIA RTX GPUs can now run some AI models locally. By using RAG on a PC, users can link to a private knowledge source – whether that be emails, notes or articles – to improve responses. The user can then feel confident that their data source, prompts and response all remain private and secure.

A recent blog provides an example of RAG accelerated by TensorRT-LLM for Windows to get better results fast.

The History of Retrieval-Augmented Generation 

The roots of the technique go back at least to the early 1970s. That’s when researchers in information retrieval prototyped what they called question-answering systems, apps that use natural language processing (NLP) to access text, initially in narrow topics such as baseball.

The concepts behind this kind of text mining have remained fairly constant over the years. But the machine learning engines driving them have grown significantly, increasing their usefulness and popularity.

In the mid-1990s, the Ask Jeeves service, now Ask.com, popularized question answering with its mascot of a well-dressed valet. IBM’s Watson became a TV celebrity in 2011 when it handily beat two human champions on the Jeopardy! game show.

Today, LLMs are taking question-answering systems to a whole new level.

Insights From a London Lab

The seminal 2020 paper arrived as Lewis was pursuing a doctorate in NLP at University College London and working for Meta at a new London AI lab. The team was searching for ways to pack more knowledge into an LLM’s parameters and using a benchmark it developed to measure its progress.

Building on earlier methods and inspired by a paper from Google researchers, the group “had this compelling vision of a trained system that had a retrieval index in the middle of it, so it could learn and generate any text output you wanted,” Lewis recalled.

When Lewis plugged into the work in progress a promising retrieval system from another Meta team, the first results were unexpectedly impressive.

“I showed my supervisor and he said, ‘Whoa, take the win. This sort of thing doesn’t happen very often,’ because these workflows can be hard to set up correctly the first time,” he said.

Lewis also credits major contributions from team members Ethan Perez and Douwe Kiela, then of New York University and Facebook AI Research, respectively.

When complete, the work, which ran on a cluster of NVIDIA GPUs, showed how to make generative AI models more authoritative and trustworthy. It’s since been cited by hundreds of papers that amplified and extended the concepts in what continues to be an active area of research.

How Retrieval-Augmented Generation Works

At a high level, here’s how an NVIDIA technical brief describes the RAG process.

When users ask an LLM a question, the AI model sends the query to another model that converts it into a numeric format so machines can read it. The numeric version of the query is sometimes called an embedding or a vector.

The embedding model then compares these numeric values to vectors in a machine-readable index of an available knowledge base. When it finds a match or multiple matches, it retrieves the related data, converts it to human-readable words and passes it back to the LLM.

Finally, the LLM combines the retrieved words and its own response to the query into a final answer it presents to the user, potentially citing sources the embedding model found.

Keeping Sources Current

In the background, the embedding model continuously creates and updates machine-readable indices, sometimes called vector databases, for new and updated knowledge bases as they become available.

Many developers find LangChain, an open-source library, can be particularly useful in chaining together LLMs, embedding models and knowledge bases. NVIDIA uses LangChain in its reference architecture for retrieval-augmented generation.

The LangChain community provides its own description of a RAG process.

Looking forward, the future of generative AI lies in creatively chaining all sorts of LLMs and knowledge bases together to create new kinds of assistants that deliver authoritative results users can verify.

Get a hands on using retrieval-augmented generation with an AI chatbot in this NVIDIA LaunchPad lab.

Artificial intelligence on Windows 11 PCs marks a pivotal moment in tech history, revolutionizing experiences for gamers, creators, streamers, office workers, students and even casual PC users.

It offers unprecedented opportunities to enhance productivity for users of the more than 100 million Windows PCs and workstations that are powered by RTX GPUs. And NVIDIA RTX technology is making it even easier for developers to create AI applications to change the way people use computers.

New optimizations, models and resources announced at Microsoft Ignite will help developers deliver new end-user experiences, quicker.

An upcoming update to TensorRT-LLM — open-source software that increases AI inference performance — will add support for new large language models and make demanding AI workloads more accessible on desktops and laptops with RTX GPUs starting at 8GB of VRAM.

TensorRT-LLM for Windows will soon be compatible with OpenAI’s popular Chat API through a new wrapper. This will enable hundreds of developer projects and applications to run locally on a PC with RTX, instead of in the cloud — so users can keep private and proprietary data on Windows 11 PCs.

Custom generative AI requires time and energy to maintain projects. The process can become incredibly complex and time-consuming, especially when trying to collaborate and deploy across multiple environments and platforms.

AI Workbench is a unified, easy-to-use toolkit that allows developers to quickly create, test and customize pretrained generative AI models and LLMs on a PC or workstation. It provides developers a single platform to organize their AI projects and tune models to specific use cases.

This enables seamless collaboration and deployment for developers to create cost-effective, scalable generative AI models quickly. Join the early access list to be among the first to gain access to this growing initiative and to receive future updates.

To support AI developers, NVIDIA and Microsoft will release are releasing DirectML enhancements to accelerate one two of the most popular foundational AI models,: Llama 2 and Stable Diffusion. Developers now have more options for cross-vendor deployment, in addition to setting a new standard for performance.

Portable AI

Last month, NVIDIA announced TensorRT-LLM for Windows, a library for accelerating LLM inference.

The next TensorRT-LLM release, v0.6.0 coming later this month, will bring improved inference performance — up to 5x faster — and enable support for additional popular LLMs, including the new Mistral 7B and Nemotron-3 8B. Versions of these LLMs will run on any GeForce RTX 30 Series and 40 Series GPU with 8GB of RAM or more, making fast, accurate, local LLM capabilities accessible even in some of the most portable Windows devices.

The new release of TensorRT-LLM will be available for install on the /NVIDIA/TensorRT-LLM GitHub repo. New optimized models will be available on ngc.nvidia.com.

Conversing With Confidence 

Developers and enthusiasts worldwide use OpenAI’s Chat API for a wide range of applications — from summarizing web content and drafting documents and emails to analyzing and visualizing data and creating presentations.

One challenge with such cloud-based AIs is that they require users to upload their input data, making them impractical for private or proprietary data or for working with large datasets.

To address this challenge, NVIDIA is soon enabling TensorRT-LLM for Windows to offer a similar API interface to OpenAI’s widely popular ChatAPI, through a new wrapper, offering a similar workflow to developers whether they are designing models and applications to run locally on a PC with RTX or in the cloud. By changing just one or two lines of code, hundreds of AI-powered developer projects and applications can now benefit from fast, local AI. Users can keep their data on their PCs and not worry about uploading datasets to the cloud.

Perhaps the best part is that many of these projects and applications are open source, making it easy for developers to leverage and extend their capabilities to fuel the adoption of generative AI on Windows, powered by RTX.

The wrapper will work with any LLM that’s been optimized for TensorRT-LLM (for example, Llama 2, Mistral and NV LLM) and is being released as a reference project on GitHub, alongside other developer resources for working with LLMs on RTX.

Model Acceleration

Developers can now leverage cutting-edge AI models and deploy with a cross-vendor API. As part of an ongoing commitment to empower developers, NVIDIA and Microsoft have been working together to accelerate Llama on RTX via the DirectML API.

Building on the announcements for the fastest inference performance for these models announced last month, this new option for cross-vendor deployment makes it easier than ever to bring AI capabilities to PC.

Developers and enthusiasts can experience the latest optimizations by downloading the latest ONNX runtime and following the installation instructions from Microsoft, and installing the latest driver from NVIDIA, which will be available on Nov. 21.

These new optimizations, models and resources will accelerate the development and deployment of AI features and applications to the 100 million RTX PCs worldwide, joining the more than 400 partners shipping AI-powered apps and games already accelerated by RTX GPUs.

As models become even more accessible and developers bring more generative AI-powered functionality to RTX-powered Windows PCs, RTX GPUs will be critical for enabling users to take advantage of this powerful technology.

Editor’s note: This post is part of our weekly In the NVIDIA Studio series, which celebrates featured artists, offers creative tips and tricks, and demonstrates how NVIDIA Studio technology improves creative workflows. We’re also deep diving on new GeForce RTX 40 Series GPU features, technologies and resources, and how they dramatically accelerate content creation.

Character animator Sir Wade Neistadt works to make animation and 3D education more accessible for aspiring and professional artists alike through video tutorials and industry training.

The YouTube creator, who goes by Sir Wade, also likes a challenge. When electronics company Razer recently asked him to create something unique and creative using the new Razer Blade 18 laptop with GeForce RTX 4090 graphics, Sir Wade obliged.

“I said yes because I thought it’d be a great opportunity to try something creatively risky and make something I didn’t yet know how to achieve,” the artist said.

I, Robot

One of the hardest parts of getting started on a project is needing to be creative on demand, said Sir Wade. For the Razer piece, the animator started by asking himself two questions: “What am I inspired by?” and “What do I have to work with?”

Sir Wade finds inspiration in games, technology, movies, people-watching and conversations. Fond of tech — and having eyed characters from the ProRigs library for some time — he decided his short animation should feature robots.

When creating a concept for the animation, Sir Wade took an unorthodox approach, skipping the popular step of 2D sketching. Instead, he captured video references by acting out the animations himself.

This gave Sir Wade the opportunity to quickly try a bunch of movements and preview body mechanics for the animation phase. Since ProRigs characters are rigs based on Autodesk Maya, he naturally began his animation work using this 3D software.

His initial approach was straightforward: mimicking the main robot character’s movements with the edited reference footage. This worked fairly well, as NVIDIA RTX-accelerated ray tracing and AI denoising with the default Autodesk Arnold renderer resulted in smooth viewport movement and photorealistic visuals.

Then, Sir Wade continued tinkering with the piece, focusing on how the robot’s arm plates crashed into each other and how its feet moved. This was a great challenge, but he kept moving on the project. The featured artist would advise, “Don’t wait for everything to be perfect.”

Next, Sir Wade exported files into Blender software with the Universal Scene Description (OpenUSD) framework, unlocking an open and extensible ecosystem, including the ability to make edits in NVIDIA Omniverse, a development platform for building and connecting 3D tools and applications. The edits could then be captured in the original native files, eliminating the need for tedious uploading, downloading and file reformatting.

Sir Wade browsed the Kitbash3D digital platform with the new asset browser Cargo to compile kits, models and materials, and drag them into Blender with ease. It’s important at this stage to get base-level models in the scene, he said, so the environment can be further refined.

Sir Wade raved about the Razer Blade 18’s quad-high-definition (QHD+) 18″ screen and 16:10 aspect ratio, which gives him more room to create, as well as its color-calibrated display, which ensures uploads to social media are as accurate as possible and require minimal color correction.

The preinstalled NVIDIA Studio Drivers, free to RTX GPU owners, are extensively tested with the most popular creative software to deliver maximum stability and performance.

Returning to the action, Sir Wade used an emission shader to form the projectiles aimed at the robot. He also tweaked various textures, such as surface imperfections, to make the robot feel more weathered and battle-worn, before moving on to visual effects (VFX).

The artist used basic primitives as particle emitters in Blender to achieve the look of bursting particles over a limited number of frames. This, combined with the robot and floor surfaces containing surface nodes, creates sparks when the robot moves or gets hit by objects.

Sir Wade’s GeForce RTX 4090 Laptop GPU with Blender Cycles RTX-accelerated OptiX ray tracing in the viewport provides interactive, photorealistic rendering for modeling and animation.

To further experiment with VFX, Sir Wade imported the project into the EmberGen simulation tool to test out various preset and physics effects.

He added dust and debris VFX, and exported the scene as an OpenVDB file back to Blender to perfect the lighting.

Finally, Sir Wade completed sound-design effects in Blackmagic Design’s DaVinci Resolve software.

Sir Wade’s video tutorials resonate with diverse audiences because of their fresh approach to solving problems and individualistic flair.

“Creativity for me doesn’t come naturally like for other artists,” Sir Wade explained. “I reverse engineer the process by seeing a tool or a concept, evaluating what’s interesting, then either figuring out a way to use it uniquely or explaining the discovery in a relatable way.”

Check out Sir Wade’s animation workshops on his website.

Less than two days remain in Sir Wade’s Fall 2023 Animation Challenge. Download the challenge template and Maya character rig files, and submit a custom 3D scene to win an NVIDIA RTX GPU or other prizes by end of day on Wednesday, Nov. 15.

Follow NVIDIA Studio on Instagram, Twitter and Facebook. Access tutorials on the Studio YouTube channel and get updates directly in your inbox by subscribing to the Studio newsletter. 

NVIDIA today unveiled at SC23 the next wave of technologies that will lift scientific and industrial research centers worldwide to new levels of performance and energy efficiency.

“NVIDIA hardware and software innovations are creating a new class of AI supercomputers,” said Ian Buck, vice president of the company’s high performance computing and hyperscale data center business, in a special address at the conference.

Some of the systems will pack memory-enhanced NVIDIA Hopper accelerators, others a new NVIDIA Grace Hopper systems architecture. All will use the expanded parallelism to run a full stack of accelerated software for generative AI, HPC and hybrid quantum computing.

Buck described the new NVIDIA HGX H200 as “the world’s leading AI computing platform.”

It packs up to 141GB of HBM3e, the first AI accelerator to use the ultrafast technology. Running models like GPT-3, NVIDIA H200 Tensor Core GPUs provide an 18x performance increase over prior-generation accelerators.

Among other generative AI benchmarks, they zip through 12,000 tokens per second on a Llama2-13B large language model (LLM).

Buck also revealed a server platform that links four NVIDIA GH200 Grace Hopper Superchips on an NVIDIA NVLink interconnect. The quad configuration puts in a single compute node a whopping 288 Arm Neoverse cores and 16 petaflops of AI performance with up to 2.3 terabytes of high-speed memory.

Demonstrating its efficiency, one GH200 Superchip using the NVIDIA TensorRT-LLM open-source library is 100x faster than a dual-socket x86 CPU system and nearly 2x more energy efficient than an X86 + H100 GPU server.

“Accelerated computing is sustainable computing,” Buck said. “By harnessing the power of accelerated computing and generative AI, together we can drive innovation across industries while reducing our impact on the environment.”

NVIDIA Powers 38 of 49 New TOP500 Systems

The latest TOP500 list of the world’s fastest supercomputers reflects the shift toward accelerated, energy-efficient supercomputing.

Thanks to new systems powered by NVIDIA H100 Tensor Core GPUs, NVIDIA now delivers more than 2.5 exaflops of HPC performance across these world-leading systems, up from 1.6 exaflops in the May rankings. NVIDIA’s contribution on the top 10 alone reaches nearly an exaflop of HPC and 72 exaflops of AI performance.

The new list contains the highest number of systems ever using NVIDIA technologies, 379 vs. 372 in May, including 38 of 49 new supercomputers on the list.

Microsoft Azure leads the newcomers with its Eagle system using H100 GPUs in NDv5 instances to hit No. 3 with 561 petaflops. Mare Nostrum5 in Barcelona ranked No. 8, and NVIDIA Eos — which recently set new AI training records on the MLPerf benchmarks — came in at No. 9.

Showing their energy efficiency, NVIDIA GPUs power 23 of the top 30 systems on the Green500. And they retained the No. 1 spot with the H100 GPU-based Henri system, which delivers 65.09 gigaflops per watt for the Flatiron Institute in New York.

Gen AI Explores COVID

Showing what’s possible, the Argonne National Laboratory used NVIDIA BioNeMo, a generative AI platform for biomolecular LLMs, to develop GenSLMs, a model that can generate gene sequences that closely resemble real-world variants of the coronavirus. Using NVIDIA GPUs and data from 1.5 million COVID genome sequences, it can also rapidly identify new virus variants.

The work won the Gordon Bell special prize last year and was trained on supercomputers, including Argonne’s Polaris system, the U.S. Department of Energy’s Perlmutter and NVIDIA’s Selene.

It’s “just the tip of the iceberg — the future is brimming with possibilities, as generative AI continues to redefine the landscape of scientific exploration,” said Kimberly Powell, vice president of healthcare at NVIDIA, in the special address.

Saving Time, Money and Energy

Using the latest technologies, accelerated workloads can see an order-of-magnitude reduction in system cost and energy used, Buck said.

For example, Siemens teamed with Mercedes to analyze aerodynamics and related acoustics for its new electric EQE vehicles. The simulations that took weeks on CPU clusters ran significantly faster using the latest NVIDIA H100 GPUs. In addition, Hopper GPUs let them reduce costs by 3x and reduce energy consumption by 4x (below).

Switching on 200 Exaflops Beginning Next Year

Scientific and industrial advances will come from every corner of the globe where the latest systems are being deployed.

“We already see a combined 200 exaflops of AI on Grace Hopper supercomputers going to production 2024,” Buck said.

They include the massive JUPITER supercomputer at Germany’s Jülich center. It can deliver 93 exaflops of performance for AI training and 1 exaflop for HPC applications, while consuming only 18.2 megawatts of power.

Based on Eviden’s BullSequana XH3000 liquid-cooled system, JUPITER will use the NVIDIA quad GH200 system architecture and NVIDIA Quantum-2 InfiniBand networking for climate and weather predictions, drug discovery, hybrid quantum computing and digital twins. JUPITER quad GH200 nodes will be configured with 864GB of high-speed memory.

It’s one of several new supercomputers using Grace Hopper that NVIDIA announced at SC23.

The HPE Cray EX2500 system from Hewlett Packard Enterprise will use the quad GH200 to power many AI supercomputers coming online next year.

For example, HPE uses the quad GH200 to power OFP-II, an advanced HPC system in Japan shared by the University of Tsukuba and the University of Tokyo, as well as the DeltaAI system, which will triple computing capacity for the U.S. National Center for Supercomputing Applications.

HPE is also building the Venado system for the Los Alamos National Laboratory, the first GH200 to be deployed in the U.S. In addition, HPE is building GH200 supercomputers in the Middle East, Switzerland and the U.K.

Grace Hopper in Texas and Beyond

At the Texas Advanced Computing Center (TACC), Dell Technologies is building the Vista supercomputer with NVIDIA Grace Hopper and Grace CPU Superchips.

More than 100 global enterprises and organizations, including NASA Ames Research Center and Total Energies, have already purchased Grace Hopper early-access systems, Buck said.

They join previously announced GH200 users such as SoftBank and the University of Bristol, as well as the massive Leonardo system with 14,000 NVIDIA A100 GPUs that delivers 10 exaflops of AI performance for Italy’s Cineca consortium.

The View From Supercomputing Centers

Leaders from supercomputing centers around the world shared their plans and work in progress with the latest systems.

“We’ve been collaborating with MeteoSwiss ECMWP as well as scientists from ETH EXCLAIM and NVIDIA’s Earth-2 project to create an infrastructure that will push the envelope in all dimensions of big data analytics and extreme scale computing,” said Thomas Schultess, director of the Swiss National Supercomputing Centre of work on the Alps supercomputer.

“There’s really impressive energy-efficiency gains across our stacks,” Dan Stanzione, executive director of TACC, said of Vista.

It’s “really the stepping stone to move users from the kinds of systems we’ve done in the past to looking at this new Grace Arm CPU and Hopper GPU tightly coupled combination and … we’re looking to scale out by probably a factor of 10 or 15 from what we are deploying with Vista when we deploy Horizon in a couple years,” he said.

Accelerating the Quantum Journey

Researchers are also using today’s accelerated systems to pioneer a path to tomorrow’s supercomputers.

In Germany, JUPITER “will revolutionize scientific research across climate, materials, drug discovery and quantum computing,” said Kristel Michelson, who leads Julich’s research group on quantum information processing.

“JUPITER’s architecture also allows for the seamless integration of quantum algorithms with parallel HPC algorithms, and this is mandatory for effective quantum HPC hybrid simulations,” she said.

CUDA Quantum Drives Progress

The special address also showed how NVIDIA CUDA Quantum — a platform for programming CPUs, GPUs and quantum computers also known as QPUs — is advancing research in quantum computing.

For example, researchers at BASF, the world’s largest chemical company, pioneered a new hybrid quantum-classical method for simulating chemicals that can shield humans against harmful metals. They join researchers at Brookhaven National Laboratory and HPE who are separately pushing the frontiers of science with CUDA Quantum.

NVIDIA also announced a collaboration with Classiq, a developer of quantum programming tools, to create a life sciences research center at the Tel Aviv Sourasky Medical Center, Israel’s largest teaching hospital.  The center will use Classiq’s software and CUDA Quantum running on an NVIDIA DGX H100 system.

Separately, Quantum Machines will deploy the first NVIDIA DGX Quantum, a system using Grace Hopper Superchips, at the Israel National Quantum Center that aims to drive advances across scientific fields. The DGX system will be connected to a superconducting QPU by Quantware and a photonic QPU from ORCA Computing, both powered by CUDA Quantum.

“In just two years, our NVIDIA quantum computing platform has amassed over 120 partners [above], a testament to its open, innovative platform,” Buck said.

Overall, the work across many fields of discovery reveals a new trend that combines accelerated computing at data center scale with NVIDIA’s full-stack innovation.

“Accelerated computing is paving the path for sustainable computing with advancements that provide not just amazing technology but a more sustainable and impactful future,” he concluded.

Watch NVIDIA’s SC23 special address below.

A widely acclaimed large language model for genomic data has demonstrated its ability to generate gene sequences that closely resemble real-world variants of SARS-CoV-2, the virus behind COVID-19.

Called GenSLMs, the model, which last year won the Gordon Bell special prize for high performance computing-based COVID-19 research, was trained on a dataset of nucleotide sequences — the building blocks of DNA and RNA. It was developed by researchers from Argonne National Laboratory, NVIDIA, the University of Chicago and a score of other academic and commercial collaborators.

When the researchers looked back at the nucleotide sequences generated by GenSLMs, they discovered that specific characteristics of the AI-generated sequences closely matched the real-world Eris and Pirola subvariants that have been prevalent this year — even though the AI was only trained on COVID-19 virus genomes from the first year of the pandemic.

“Our model’s generative process is extremely naive, lacking any specific information or constraints around what a new COVID variant should look like,” said Arvind Ramanathan, lead researcher on the project and a computational biologist at Argonne. “The AI’s ability to predict the kinds of gene mutations present in recent COVID strains — despite having only seen the Alpha and Beta variants during training — is a strong validation of its capabilities.”

In addition to generating its own sequences, GenSLMs can also classify and cluster different COVID genome sequences by distinguishing between variants. In a demo coming soon to NGC, NVIDIA’s hub for accelerated software, users can explore visualizations of GenSLMs’ analysis of the evolutionary patterns of various proteins within the COVID viral genome.

Reading Between the Lines, Uncovering Evolutionary Patterns

A key feature of GenSLMs is its ability to interpret long strings of nucleotides — represented with sequences of the letters A, T, G and C in DNA, or A, U, G and C in RNA — in the same way an LLM trained on English text would interpret a sentence. This capability enables the model to understand the relationship between different areas of the genome, which in coronaviruses consists of around 30,000 nucleotides.

In the demo, users will be able to choose from among eight different COVID variants to understand how the AI model tracks mutations across various proteins of the viral genome. The visualization depicts evolutionary couplings across the viral proteins — highlighting which snippets of the genome are likely to be seen in a given variant.

“Understanding how different parts of the genome are co-evolving gives us clues about how the virus may develop new vulnerabilities or new forms of resistance,” Ramanathan said. “Looking at the model’s understanding of which mutations are particularly strong in a variant may help scientists with downstream tasks like determining how a specific strain can evade the human immune system.”

GenSLMs was trained on more than 110 million prokaryotic genome sequences and fine-tuned with a global dataset of around 1.5 million COVID viral sequences using open-source data from the Bacterial and Viral Bioinformatics Resource Center. In the future, the model could be fine-tuned on the genomes of other viruses or bacteria, enabling new research applications.

To train the model, the researchers used NVIDIA A100 Tensor Core GPU-powered supercomputers, including Argonne’s Polaris system, the U.S. Department of Energy’s Perlmutter and NVIDIA’s Selene.

The GenSLMs research team’s Gordon Bell special prize was awarded at last year’s SC22 supercomputing conference. At this week’s SC23, in Denver, NVIDIA is sharing a new range of groundbreaking work in the field of accelerated computing. View the full schedule.

NVIDIA Research comprises hundreds of scientists and engineers worldwide, with teams focused on topics including AI, computer graphics, computer vision, self-driving cars and robotics. Learn more about NVIDIA Research and subscribe to NVIDIA healthcare news.

Main image courtesy of Argonne National Laboratory’s Bharat Kale. 

This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. DOE Office of Science and the National Nuclear Security Administration. Research was supported by the DOE through the National Virtual Biotechnology Laboratory, a consortium of DOE national laboratories focused on response to COVID-19, with funding from the Coronavirus CARES Act.

Michael Kuehn and Davide Vodola are taking to new heights work that’s pioneering quantum computing for the world’s largest chemical company.

The BASF researchers are demonstrating how a quantum algorithm can see what no traditional simulation can — key attributes of NTA, a compound with applications that include removing toxic metals like iron from a city’s wastewater.

The quantum computing team at BASF simulated on GPUs how the equivalent of 24 qubits — the processing engines of a quantum computer — can tackle the challenge.

Many corporate R&D centers would consider that a major achievement, but they pressed on, and recently ran their first 60 qubit simulations on NVIDIA’s Eos H100 Supercomputer.

“It’s the largest simulation of a molecule using a quantum algorithm we’ve ever run,” said Kuehn.

Flexible, Friendly Software

BASF is running the simulation on NVIDIA CUDA Quantum, a platform for programming CPUs, GPUs and quantum computers, also known as QPUs.

Vodola described it as “very flexible and user friendly, letting us build up a complex quantum circuit simulation from relatively simple building blocks. Without CUDA Quantum, it would be impossible to run this simulation,” he said.

The work requires a lot of heavy lifting, too, so BASF turned to an NVIDIA DGX Cloud service that uses NVIDIA H100 Tensor Core GPUs.

“We need a lot of computing power, and the NVIDIA platform is significantly faster than CPU-based hardware for this kind of simulation,” said Kuehn.

BASF’s quantum computing initiative, which Kuehn helped launch, started in 2017. In addition to its work in chemistry, the team is developing use cases for quantum computing in machine learning as well as optimizations for logistics and scheduling.

An Expanding CUDA Quantum Community

Other research groups are also advancing science with CUDA Quantum.

At SUNY Stony Brook, researchers are pushing the boundaries of high-energy physics to simulate complex interactions of subatomic particles. Their work promises new discoveries in fundamental physics.

“CUDA Quantum enables us to do quantum simulations that would otherwise be impossible,” said Dmitri Kharzeev,  a SUNY professor and scientist at Brookhaven National Lab.

In addition, a research team at Hewlett Packard Labs is using the Perlmutter supercomputer to explore magnetic phase transition in quantum chemistry in one of the largest simulations of its kind. The effort could reveal important and unknown details of physical processes too difficult to model with conventional techniques.

“As quantum computers progress toward useful applications, high-performance classical simulations will be key for prototyping novel quantum algorithms,” said Kirk Bresniker, a chief architect at Hewlett Packard Labs. “Simulating and learning from quantum data are promising avenues toward tapping quantum computing’s potential.”

A Quantum Center for Healthcare

These efforts come as support for CUDA Quantum expands worldwide.

Classiq — an Israeli startup that already has more than 400 universities using its novel approach to writing quantum programs — announced today a new research center at the Tel Aviv Sourasky Medical Center, Israel’s largest teaching hospital.

Created in collaboration with NVIDIA, it will train experts in life science to write quantum applications that could someday help doctors diagnose diseases or accelerate the discovery of new drugs.

Classiq created quantum design software that automates low-level tasks, so developers don’t need to know all the complex details of how a quantum computer works. It’s now being integrated with CUDA Quantum.

Terra Quantum, a quantum services company with headquarters in Germany and Switzerland, is developing hybrid quantum applications for life sciences, energy, chemistry and finance that will run on CUDA Quantum. And IQM in Finland is enabling its superconducting QPU to use CUDA Quantum.

Quantum Loves Grace Hopper

Several companies, including Oxford Quantum Circuits, will use NVIDIA Grace Hopper Superchips to power their hybrid quantum efforts. Based in Reading, England, Oxford Quantum is using Grace Hopper in a hybrid QPU/GPU system programmed by CUDA Quantum.

Quantum Machines announced that the Israeli National Quantum Center will be the first deployment of NVIDIA DGX Quantum, a system using Grace Hopper Superchips. Based in Tel Aviv, the center will tap DGX Quantum to power quantum computers from Quantware, ORCA Computing and more.

In addition, Grace Hopper is being put to work by qBraid, in Chicago, to build a quantum cloud service, and Fermioniq, in Amsterdam, to develop tensor-network algorithms.

The large quantity of shared memory and the memory bandwidth of Grace Hopper make these superchips an excellent fit for memory-hungry quantum simulations.

Get started programming hybrid quantum systems today with the latest release of CUDA Quantum from NGC, NVIDIA’s catalog of accelerated software, or GitHub.

Dozens of new supercomputers for scientific computing will soon hop online, powered by NVIDIA’s breakthrough GH200 Grace Hopper Superchip for giant-scale AI and high performance computing.

The NVIDIA GH200 enables scientists and researchers to tackle the world’s most challenging problems by accelerating complex AI and HPC applications running terabytes of data.

At the SC23 supercomputing show, NVIDIA today announced that the superchip is coming to more systems worldwide, including from Dell Technologies, Eviden, Hewlett Packard Enterprise (HPE), Lenovo, QCT and Supermicro.

Bringing together the Arm-based NVIDIA Grace CPU and Hopper GPU architectures using NVIDIA NVLink-C2C interconnect technology, GH200 also serves as the engine behind scientific supercomputing centers across the globe.

Combined, these GH200-powered centers represent some 200 exaflops of AI performance to drive scientific innovation.

HPE Cray Supercomputers Integrate NVIDIA Grace Hopper

At the show in Denver, HPE announced it will offer HPE Cray EX2500 supercomputers with the NVIDIA Grace Hopper Superchip. The integrated solution will feature quad GH200 processors, scaling up to tens of thousands of Grace Hopper Superchip nodes to provide organizations with unmatched supercomputing agility and quicker AI training. This configuration will also be part of a supercomputing solution for generative AI that HPE introduced today.

“Organizations are rapidly adopting generative AI to accelerate business transformations and technological breakthroughs,” said Justin Hotard, executive vice president and general manager of HPC, AI and Labs at HPE. “Working with NVIDIA, we’re excited to deliver a full supercomputing solution for generative AI, powered by technologies like Grace Hopper, which will make it easy for customers to accelerate large-scale AI model training and tuning at new levels of efficiency.”

Next-Generation AI Supercomputing Centers

A vast array of the world’s supercomputing centers are powered by NVIDIA Grace Hopper systems. Several top centers announced at SC23 that they’re now integrating GH200 systems for their supercomputers.

Germany’s Jülich Supercomputing Centre will use GH200 superchips in JUPITER, set to become the first exascale supercomputer in Europe. The supercomputer will help tackle urgent scientific challenges, such as mitigating climate change, combating pandemics and bolstering sustainable energy production.

Japan’s Joint Center for Advanced High Performance Computing — established between the Center for Computational Sciences at the University of Tsukuba and the Information Technology Center at the University of Tokyo — promotes advanced computational sciences integrated with data analytics, AI and machine learning across academia and industry. Its next-generation supercomputer will be powered by NVIDIA Grace Hopper.

The Texas Advanced Computing Center, based in Austin, Texas, designs and operates some of the world’s most powerful computing resources. The center will power its Vista supercomputer with NVIDIA GH200 for low power and high-bandwidth memory to deliver more computation while enabling bigger models to run with greater efficiency.

The National Center for Supercomputing Applications at the University of Illinois Urbana-Champaign will tap NVIDIA Grace Hopper superchips to power DeltaAI, an advanced computing and data resource set to triple NCSA’s AI-focused computing capacity.

And, the University of Bristol recently received funding from the UK government to build Isambard-AI, set to be the country’s most powerful supercomputer, which will enable AI-driven breakthroughs in robotics, big data, climate research and drug discovery. The new system, being built by HPE, will be equipped with over 5,000 NVIDIA GH200 Grace Hopper Superchips, providing 21 exaflops of AI supercomputing power capable of making 21 quintillion AI calculations per second.

These systems join previously announced next-generation Grace Hopper systems from the Swiss National Supercomputing Centre, Los Alamos National Laboratory and SoftBank Corp.

GH200 Shipping Globally and Available in Early Access from CSPs

GH200 is available in early access from select cloud service providers such as Lambda and Vultr. Oracle Cloud Infrastructure today announced plans to offer GH200 instances, while CoreWeave detailed plans for early availability of its GH200 instances starting in Q1 2024.

Other system manufacturers such as ASRock Rack, ASUS, GIGABYTE and Ingrasys will begin shipping servers with the superchips by the end of the year.

NVIDIA Grace Hopper has been adopted in early access for supercomputing initiatives by more than 100 enterprises, organizations and government agencies across the globe, including the NASA Ames Research Center for aeronautics research and global energy company TotalEnergies.

In addition, the GH200 will soon become available through NVIDIA LaunchPad, which provides free access to enterprise NVIDIA hardware and software through an internet browser.

Learn more about Grace Hopper and other supercomputing breakthroughs by joining NVIDIA at SC23.

Thanks to a viral trend sweeping social media, we now know some men think about the Roman Empire every day.

And thanks to Luke Farritor, a 21-year-old computer science undergrad at the University of Nebraska-Lincoln, and like-minded AI enthusiasts, there might soon be a lot more to think about.

Blending a passion for history with machine learning skills, Farritor has triumphed in the Vesuvius Challenge, wielding the power of the NVIDIA GeForce GTX 1070 GPU to bring a snippet of ancient text back from the ashes after almost 2,000 years.

Text Big Thing: Deciphering Rome’s Hidden History

The Herculaneum scrolls are a library of ancient texts that were carbonized and preserved by the eruption of Mount Vesuvius in 79 AD, which buried the cities of Pompeii and Herculaneum under a thick layer of ash and pumice.

The competition, which has piqued the interest of historians and technologists across the globe, seeks to extract readable content from the carbonized remains of the scrolls.

In a significant breakthrough, the word “πορφυρας,” which means “purple dye” or “cloths of purple,” emerged from the ancient texts thanks to the efforts of Farritor.

His achievement in identifying 10 letters within a small patch of scroll earned him a $40,000 prize.

Close on his heels was Youssef Nader, a biorobotics graduate student, who independently discerned the same word a few months later, meriting a $10,000 prize.

Adding to these notable successes, Casey Handmer, an entrepreneur with a keen eye, secured another $10,000 for his demonstration that significant amounts of ink were waiting to be discovered within the unopened scrolls.

All these discoveries are advancing the work pioneered by W. Brent Seales, chair of the University of Kentucky Computer Science Department, who has dedicated over a decade to developing methods to digitally unfurl and read the delicate Herculaneum scrolls.

Turbocharging these efforts is Nat Friedman, the CEO of GitHub and the organizer of the Vesuvius Challenge, whose commitment to open-source innovation has fostered a community where such historical breakthroughs are possible.

To become the first to decipher text from the scrolls, Farritor, who served as an intern at SpaceX, harnessed the GeForce GTX 1070 to accelerate his work.

When Rome Meets RAM: Older GPU Helps Uncover Even Older Text

Introduced in 2016, the GTX 1070 is celebrated among gamers, who have long praised the GPU for its balance of performance and affordability.

Instead of gaming, however, Farritor harnessed the parallel processing capabilities of the GPU to accelerate the ResNet deep learning framework, processing data at speeds unattainable by traditional computing methods.

Farritor is not the only competitor harnessing NVIDIA GPUs, which have proven themselves as indispensable tools to Vesuvius challenge competitors.

Latin Lingo and Lost Text

Discovered in the 18th century in the Villa of the Papyri, the Herculaneum scrolls have presented a challenge to researchers. Their fragile state has made them nearly impossible to read without causing damage. The advent of advanced imaging and AI technology changed that.

The project has become a passion for Farritor, who finds himself struggling to recall more of the Latin he studied in high school. “And man, like what’s in the scrolls … it’s just the anticipation, you know?” Farritor said.

The next challenge is to unearth passages from the Herculaneum scrolls that are 144 characters long, echoing the brevity of an original Twitter post.

Engaging over 1,500 experts in a collaborative effort, the endeavor is now more heated than ever.

Private donors have upped the ante, offering a $700,000 prize for those who can retrieve four distinct passages of at least 140 characters this year — a testament to the value placed on these ancient texts and the lengths required to reclaim them.

And Farritor’s eager to keep digging, reeling off the names of lost works of Roman and Greek history that he’d love to help uncover.

He reports he’s now thinking about Rome — and what his efforts might help discover — not just every day now, but “every hour.” “I think anything that sheds light on that time in human history is gonna be significant,” Farritor said.

Wield the blade and embrace the way of the samurai for some thrilling action — Onimusha: Warlords comes to GeForce NOW this week. Members can experience feudal Japan in this hack-and-slash adventure game in the cloud.

It’s part of an action-packed GFN Thursday, with 16 more games joining the cloud gaming platform’s library.

Forging Destinies

Capcom’s popular Onimusha: Warlords is newly supported in the cloud this week, just in time for those tuning into the recently released Netflix anime adaptation.

Fight against the evil warlord Nobunaga Oda and his army of demons as samurai Samanosuke Akechi. Explore feudal Japan, wield swords, use ninja techniques and solve puzzles to defeat enemies. The action-adventure hack-and-slash game has been enhanced with improved controls for smoother swordplay mechanics, an updated soundtrack and more.

Ultimate members can stream the game in ultrawide resolution with up to eight hours each gaming session for riveting samurai action.

Endless Games

Roguelite fans and GeForce NOW members have been enjoying Sega’s Endless Dungeon in the cloud. Recruit a team of shipwrecked heroes, plunge into a long-abandoned space station and protect the crystal against never-ending waves of monsters. Never accept defeat — get reloaded and try, try again.

On top of that, check out the 16 newly supported games joining the GeForce NOW library this week:

• The Invincible (New release on Steam, Nov. 6)
• Roboquest (New release on Steam, Nov. 7)
• Stronghold: Definitive Edition (New release on Steam, Nov. 7)
• Dungeons 4 (New release on Steam, Xbox and available on PC Game Pass, Nov. 9)
• Space Trash Scavenger (New release on Steam, Nov. 9)
• Airport CEO (Steam)
• Car Mechanic Simulator 2021 (Xbox, available on PC Game Pass)
• Farming Simulator 19 (Xbox, available on Microsoft Store)
• GoNNER (Xbox, available on Microsoft Store)
• GoNNER2 (Xbox, available on Microsoft Store)
• Jurassic World Evolution 2 (Xbox, available on PC Game Pass)
• Onimusha: Warlords (Steam)
• Planet of Lana (Xbox, available on PC Game Pass)
• Q.U.B.E. 10th Anniversary (Epic Games Store)
• Trailmakers (Xbox, available on PC Game Pass)
• Turnip Boy Commits Tax Evasion (Epic Games Store)

What are you planning to play this weekend? Let us know on Twitter or in the comments below.

Share a game that more people should be playing.

— NVIDIA GeForce NOW (@NVIDIAGFN) November 8, 2023

NVIDIA’s AI platform raised the bar for AI training and high performance computing in the latest MLPerf industry benchmarks.

Among many new records and milestones, one in generative AI stands out: NVIDIA Eos — an AI supercomputer powered by a whopping 10,752 NVIDIA H100 Tensor Core GPUs and NVIDIA Quantum-2 InfiniBand networking — completed a training benchmark based on a GPT-3 model with 175 billion parameters trained on one billion tokens in just 3.9 minutes.

That’s a nearly 3x gain from 10.9 minutes, the record NVIDIA set when the test was introduced less than six months ago.

The benchmark uses a portion of the full GPT-3 data set behind the popular ChatGPT service that, by extrapolation, Eos could now train in just eight days, 73x faster than a prior state-of-the-art system using 512 A100 GPUs.

The acceleration in training time reduces costs, saves energy and speeds time-to-market. It’s heavy lifting that makes large language models widely available so every business can adopt them with tools like NVIDIA NeMo, a framework for customizing LLMs.

In a new generative AI test ‌this round, 1,024 NVIDIA Hopper architecture GPUs completed a training benchmark based on the Stable Diffusion text-to-image model in 2.5 minutes, setting a high bar on this new workload.

By adopting these two tests, MLPerf reinforces its leadership as the industry standard for measuring AI performance, since generative AI is the most transformative technology of our time.

System Scaling Soars

The latest results were due in part to the use of the most accelerators ever applied to an MLPerf benchmark. The 10,752 H100 GPUs far surpassed the scaling in AI training in June, when NVIDIA used 3,584 Hopper GPUs.

The 3x scaling in GPU numbers delivered a 2.8x scaling in performance, a 93% efficiency rate thanks in part to software optimizations.

Efficient scaling is a key requirement in generative AI because LLMs are growing by an order of magnitude every year. The latest results show NVIDIA’s ability to meet this unprecedented challenge for even the world’s largest data centers.

The achievement is thanks to a full-stack platform of innovations in accelerators, systems and software that both Eos and Microsoft Azure used in the latest round.

Eos and Azure both employed 10,752 H100 GPUs in separate submissions. They achieved within 2% of the same performance, demonstrating the efficiency of NVIDIA AI in data center and public-cloud deployments.

NVIDIA relies on Eos for a wide array of critical jobs. It helps advance initiatives like NVIDIA DLSS, AI-powered software for state-of-the-art computer graphics and NVIDIA Research projects like ChipNeMo, generative AI tools that help design next-generation GPUs.

Advances Across Workloads

NVIDIA set several new records in this round in addition to making advances in generative AI.

For example, H100 GPUs were 1.6x faster than the prior-round training recommender models widely employed to help users find what they’re looking for online. Performance was up 1.8x on RetinaNet, a computer vision model.

These increases came from a combination of advances in software and scaled-up hardware.

NVIDIA was once again the only company to run all MLPerf tests. H100 GPUs demonstrated the fastest performance and the greatest scaling in each of the nine benchmarks.

Speedups translate to faster time to market, lower costs and energy savings for users training massive LLMs or customizing them with frameworks like NeMo for the specific needs of their business.

Eleven systems makers used the NVIDIA AI platform in their submissions this round, including ASUS, Dell Technologies, Fujitsu, GIGABYTE, Lenovo, QCT and Supermicro.

NVIDIA partners participate in MLPerf because they know it’s a valuable tool for customers evaluating AI platforms and vendors.

HPC Benchmarks Expand

In MLPerf HPC, a separate benchmark for AI-assisted simulations on supercomputers, H100 GPUs delivered up to twice the performance of NVIDIA A100 Tensor Core GPUs in the last HPC round. The results showed up to 16x gains since the first MLPerf HPC round in 2019.

The benchmark included a new test that trains OpenFold, a model that predicts the 3D structure of a protein from its sequence of amino acids. OpenFold can do in minutes vital work for healthcare that used to take researchers weeks or months.

Understanding a protein’s structure is key to finding effective drugs fast because most drugs act on proteins, the cellular machinery that helps control many biological processes.

In the MLPerf HPC test, H100 GPUs trained OpenFold in 7.5 minutes.  The OpenFold test is a representative part of the entire AlphaFold training process that two years ago took 11 days using 128 accelerators.

A version of the OpenFold model and the software NVIDIA used to train it will be available soon in NVIDIA BioNeMo, a generative AI platform for drug discovery.

Several partners made submissions on the NVIDIA AI platform in this round. They included Dell Technologies and supercomputing centers at Clemson University, the Texas Advanced Computing Center and — with assistance from Hewlett Packard Enterprise (HPE) — Lawrence Berkeley National Laboratory.

Benchmarks With Broad Backing

Since its inception in May 2018, the MLPerf benchmarks have enjoyed broad backing from both industry and academia. Organizations that support them include Amazon, Arm, Baidu, Google, Harvard, HPE, Intel, Lenovo, Meta, Microsoft, NVIDIA, Stanford University and the University of Toronto.

MLPerf tests are transparent and objective, so users can rely on the results to make informed buying decisions.

All the software NVIDIA used is available from the MLPerf repository, so all developers can get the same world-class results. These software optimizations get continuously folded into containers available on NGC, NVIDIA’s software hub for GPU applications.

Learn more about MLPerf and the details of this round.