NVIDIA Hopper Leaps Ahead in Generative AI at MLPerf
27/03/2024
In the latest MLPerf benchmarks, NVIDIA TensorRT-LLM - software that speeds and simplifies the complex job of inference on large language models - boosted the performance of NVIDIA Hopper architecture GPUs on the GPT-J LLM nearly 3x over their results just six months ago.
The dramatic speedup demonstrates the power of NVIDIA's full-stack platform of chips, systems and software to handle the demanding requirements of running generative AI.
Leading companies are using TensorRT-LLM to optimize their models. And NVIDIA NIM - a set of inference microservices that includes inferencing engines like TensorRT-LLM - makes it easier than ever for businesses to deploy NVIDIA's inference platform.
Raising the Bar in Generative AI TensorRT-LLM running on NVIDIA H200 Tensor Core GPUs - the latest, memory-enhanced Hopper GPUs - delivered the fastest performance running inference in MLPerf's biggest test of generative AI to date.
The new benchmark uses the largest version of Llama 2, a state-of-the-art large language model packing 70 billion parameters. The model is more than 10x larger than the GPT-J LLM first used in the September benchmarks.
The memory-enhanced H200 GPUs, in their MLPerf debut, used TensorRT-LLM to produce up to 31,000 tokens/second, a record on MLPerf's Llama 2 benchmark.
The H200 GPU results include up to 14% gains from a custom thermal solution. It's one example of innovations beyond standard air cooling that systems builders are applying to their NVIDIA MGX designs to take the performance of Hopper GPUs to new heights.
Memory Boost for NVIDIA Hopper GPUs NVIDIA is sampling H200 GPUs to customers today and shipping in the second quarter. They'll be available soon from nearly 20 leading system builders and cloud service providers.
H200 GPUs pack 141GB of HBM3e running at 4.8TB/s. That's 76% more memory flying 43% faster compared to H100 GPUs. These accelerators plug into the same boards and systems and use the same software as H100 GPUs.
With HBM3e memory, a single H200 GPU can run an entire Llama 2 70B model with the highest throughput, simplifying and speeding inference.
GH200 Packs Even More Memory Even more memory - up to 624GB of fast memory, including 144GB of HBM3e - is packed in NVIDIA GH200 Superchips, which combine on one module a Hopper architecture GPU and a power-efficient NVIDIA Grace CPU. NVIDIA accelerators are the first to use HBM3e memory technology.
With nearly 5 TB/second memory bandwidth, GH200 Superchips delivered standout performance, including on memory-intensive MLPerf tests such as recommender systems.
Sweeping Every MLPerf Test On a per-accelerator basis, Hopper GPUs swept every test of AI inference in the latest round of the MLPerf industry benchmarks.
In addition, NVIDIA Jetson Orin remains at the forefront in MLPerf's edge category. In the last two inference rounds, Orin ran the most diverse set of models in the category, including GPT-J and Stable Diffusion XL.
The MLPerf benchmarks cover today's most popular AI workloads and scenarios, including generative AI, recommendation systems, natural language processing, speech and computer vision. NVIDIA was the only company to submit results on every workload in the latest round and every round since MLPerf's data center inference benchmarks began in October 2020.
Continued performance gains translate into lower costs for inference, a large and growing part of the daily work for the millions of NVIDIA GPUs deployed worldwide.
Advancing What's Possible Pushing the boundaries of what's possible, NVIDIA demonstrated three innovative techniques in a special section of the benchmarks called the open division, created for testing advanced AI methods.
NVIDIA engineers used a technique called structured sparsity - a way of reducing calculations, first introduced with NVIDIA A100 Tensor Core GPUs - to deliver up to 33% speedups on inference with Llama 2.
A second open division test found inference speedups of up to 40% using pruning, a way of simplifying an AI model - in this case, an LLM - to increase inference throughput.
Finally, an optimization called DeepCache reduced the math required for inference with the Stable Diffusion XL model, accelerating performance by a whopping 74%.
All these results were run on NVIDIA H100 Tensor Core GPUs.
A Trusted Source for Users MLPerf's tests are transparent and objective, so users can rely on the results to make informed buying decisions.
NVIDIA's partners participate in MLPerf because they know it's a valuable tool for customers evaluating AI systems and services. Partners submitting results on the NVIDIA AI platform in this round included ASUS, Cisco, Dell Technologies, Fujitsu, GIGABYTE, Google, Hewlett Packard Enterprise, Lenovo, Microsoft Azure, Oracle, QCT, Supermicro, VMware (recently acquired by Broadcom) and Wiwynn.
All the software NVIDIA used in the tests is available in the MLPerf repository. These optimizations are continuously folded into containers available on NGC, NVIDIA's software hub for GPU applications, as well as NVIDIA AI Enterprise - a secure, supported platform that includes NIM inference microservices.
The Next Big Thing The use cases, model sizes and datasets for generative AI continue to expand. That's why MLPerf continues to evolve, adding real-world tests with popular models like Llama 2 70B and Stable Diffusion XL.
Keeping pace with the explosion in LLM model sizes, NVIDIA founder and CEO Jensen Huang announced last week at GTC that the NVIDIA Blackwell architecture GPUs will deliver new levels of performance required for the multitrillion-parameter AI models.
Inference for large language models is difficult, requiring both e
More from Nvidia
25/04/2024
AI Drives Future of Transportation at Asia's Largest Automotive Show
The latest trends and technologies in the automotive industry are in the spotlight at the Beijing International Automotive Exhibition, aka Auto China, which ope...
25/04/2024
Into the Omniverse: Unlocking the Future of Manufacturing With OpenUSD on Siemens Teamcenter X
Editor's note: This post is part of Into the Omniverse, a series focused on ...
25/04/2024
Blast From the Past: Stream StarCraft' and Diablo' on GeForce NOW
Support for Battle.net on GeForce NOW expands this GFN Thursday, as titles from the iconic StarCraft and Diablo series come to the cloud. StarCraft Remastered,...
24/04/2024
Rays Up: Decoding AI-Powered DLSS 3.5 Ray Reconstruction
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and which showcases new hardware, ...
24/04/2024
Forecasting the Future: AI2's Christopher Bretherton Discusses Using Machine Learning for Climate Modeling
Can machine learning help predict extreme weather events and climate change? Chr...
24/04/2024
NVIDIA to Acquire GPU Orchestration Software Provider Run:ai
To help customers make more efficient use of their AI computing resources, NVIDIA today announced it has entered into a definitive agreement to acquire Run:ai, ...
24/04/2024
How Virtual Factories Are Making Industrial Digitalization a Reality
To address the shift to electric vehicles, increased semiconductor demand, manufacturing onshoring, and ambitions for greater sustainability, manufacturers are ...
23/04/2024
Small and Mighty: NVIDIA Accelerates Microsoft's Open Phi-3 Mini Language Models
NVIDIA announced today its acceleration of Microsoft's new Phi-3 Mini open l...
22/04/2024
Climate Tech Startups Integrate NVIDIA AI for Sustainability Applications
Whether they're monitoring miniscule insects or delivering insights from satellites in space, NVIDIA-accelerated startups are making every day Earth Day. S...
18/04/2024
Wide Open: NVIDIA Accelerates Inference on Meta Llama 3
NVIDIA today announced optimizations across all its platforms to accelerate Meta Llama 3, the latest generation of the large language model (LLM). The open mod...
18/04/2024
Up to No Good: No Rest for the Wicked' Early Access Launches on GeForce NOW
It's time to get a little wicked. Members can now stream No Rest for the Wicked from the cloud. It leads six new games joining the GeForce NOW library of m...
18/04/2024
NVIDIA Honors Partners of the Year in Europe, Middle East, Africa
NVIDIA today recognized 18 partners in Europe, the Middle East and Africa for their achievements and commitment to driving AI adoption. The recipients were hon...
17/04/2024
Seeing Beyond: Living Optics CEO Robin Wang on Democratizing Hyperspectral Imaging
Step into the realm of the unseen with Robin Wang, CEO of Living Optics. The sta...
17/04/2024
Moving Pictures: Transform Images Into 3D Scenes With NVIDIA Instant NeRF
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and which showcases new hardware, ...
16/04/2024
New NVIDIA RTX A400 and A1000 GPUs Enhance AI-Powered Design and Productivity Workflows
AI integration across design and productivity applications is becoming the new s...
16/04/2024
To Cut a Long Story Short: Video Editors Benefit From DaVinci Resolve's New AI Features Powered by RTX
Editor's note: This post is part of our In the NVIDIA Studio series, which c...
15/04/2024
AI Is Tech's Greatest Contribution to Social Elevation,' NVIDIA CEO Tells Oregon State Students
AI promises to bring the full benefits of the digital revolution to billions acr...
10/04/2024
The Building Blocks of AI: Decoding the Role and Significance of Foundation Models
Editor's note: This post is part of the AI Decoded series, which demystifies...
10/04/2024
Combating Corruption With Data: Cleanlab and Berkeley Research Group on Using AI-Powered Investigative Analytics
Talk about scrubbing data. Curtis Northcutt, cofounder and CEO of Cleanlab, and ...
09/04/2024
NVIDIA Joins $110 Million Partnership to Help Universities Teach AI Skills
The Biden Administration has announced a new $110 million AI partnership between Japan and the United States that includes an initiative to fund research throug...
09/04/2024
Broadcasting Breakthroughs: NVIDIA Holoscan for Media, Available Now, Transforms Live Media With Easy AI Integration
Whether delivering live sports programming, streaming services, network broadcas...
09/04/2024
Start Up Your Engines: NVIDIA and Google Cloud Collaborate to Accelerate AI Development
NVIDIA and Google Cloud have announced a new collaboration to help startups arou...
04/04/2024
NVIDIA Ranked by Fortune at No. 3 on 100 Best Companies to Work For' List
NVIDIA jumped to No. 3 on the latest list of America's 100 Best Companies to Work For by Fortune magazine and Great Place to Work. It's the company'...
04/04/2024
The Elder Scrolls Online' Joins GeForce NOW for Game's 10th Anniversary
Rain or shine, a new month means new games. GeForce NOW kicks off April with nearly 20 new games, seven of which are available to play this week. GFN Thursday ...
03/04/2024
A New Lens: Dotlumen CEO Cornel Amariei on Assistive Technology for the Visually Impaired
Dotlumen is illuminating a new technology to help people with visual impairments...
03/04/2024
Coming Up ACEs: Decoding the AI Technology That's Enhancing Games With Realistic Digital Humans
Editor's note: This post is part of the AI Decoded series, which demystifies...
28/03/2024
Greater Scope: Doctors Get Inside Look at Gut Health With AI-Powered Endoscopy
From humble beginnings as a university spinoff to an acquisition by the leading global medtech company in its field, Odin Vision has been on an accelerated jour...
28/03/2024
Get Cozy With Palia' on GeForce NOW
Ease into spring with the warm, cozy vibes of Palia, coming to the cloud this GFN Thursday. It's part of six new titles joining the GeForce NOW library of ...
27/03/2024
Software Developers Launch OpenUSD and Generative AI-Powered Product Configurators Built on NVIDIA Omniverse
From designing dream cars to customizing clothing, 3D product configurators are ...
27/03/2024
NVIDIA Hopper Leaps Ahead in Generative AI at MLPerf
It's official: NVIDIA delivered the world's fastest platform in industry-standard tests for inference on generative AI. In the latest MLPerf benchmarks...
27/03/2024
Viome's Guru Banavar Discusses AI for Personalized Health
In the latest episode of NVIDIA's AI Podcast, Viome Chief Technology Officer Guru Banavar spoke with host Noah Kravitz about how AI and RNA sequencing are r...
27/03/2024
Unlocking Peak Generations: TensorRT Accelerates AI on RTX PCs and Workstations
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and which showcases new hardware, ...
26/03/2024
Boom in AI-Enabled Medical Devices Transforms Healthcare
The future of healthcare is software-defined and AI-enabled. Around 700 FDA-cleared, AI-enabled medical devices are now on the market - more than 10x the number...
26/03/2024
Model Innovators: How Digital Twins Are Making Industries More Efficient
A manufacturing plant near Hsinchu, Taiwan's Silicon Valley, is among facilities worldwide boosting energy efficiency with AI-enabled digital twins. A virt...
26/03/2024
Into the Omniverse: Groundbreaking OpenUSD Advancements Put NVIDIA GTC Spotlight on Developers
Editor's note: This post is part of Into the Omniverse, a series focused on ...
25/03/2024
NVIDIA Blackwell and Automotive Industry Innovators Dazzle at NVIDIA GTC
Generative AI, in the data center and in the car, is making vehicle experiences safer and more enjoyable. The latest advancements in automotive technology were...
21/03/2024
AI's New Frontier: From Daydreams to Digital Deeds
Imagine a world where you can whisper your digital wishes into your device, and poof, it happens. That world may be coming sooner than you think. But if you...
21/03/2024
You Transformed the World,' NVIDIA CEO Tells Researchers Behind Landmark AI Paper
Of GTC's 900+ sessions, the most wildly popular was a conversation hosted by...
21/03/2024
Instant Latte: NVIDIA Gen AI Research Brews 3D Shapes in Under a Second
NVIDIA researchers have pumped a double shot of acceleration into their latest text-to-3D generative AI model, dubbed LATTE3D. Like a virtual 3D printer, LATTE...
21/03/2024
Here Be Dragons: Dragon's Dogma 2' Comes to GeForce NOW
Arise for a new adventure with Dragon's Dogma 2, leading two new titles joining the GeForce NOW library this week. Set Forth, Arisen Fulfill a forgotten de...
20/03/2024
AI Decoded From GTC: The Latest Developer Tools and Apps Accelerating AI on PC and Workstation
Editor's note: This post is part of the AI Decoded series, which demystifies...
19/03/2024
NVIDIA Celebrates Americas Partners Driving AI-Powered Transformation
NVIDIA recognized 14 partners in the Americas for their achievements in transforming businesses with AI, this week at GTC. The winners of the NVIDIA Partner Ne...
19/03/2024
Climate Pioneers: 3 Startups Harnessing NVIDIA's AI and Earth-2 Platforms
To help mitigate climate change - one of humanity's greatest challenges - researchers are turning to AI and sustainable computing to accelerate and operatio...
19/03/2024
Secure by Design: NVIDIA AIOps Partner Ecosystem Blends AI for Businesses
In today's complex business environments, IT teams face a constant flow of challenges, from simple issues like employee account lockouts to critical securit...
19/03/2024
Generation Sensation: New Generative AI and RTX Tools Boost Content Creation
Editor's note: This post is part of our In the NVIDIA Studio series, which celebrates featured artists, offers creative tips and tricks, and demonstrates ho...
19/03/2024
NVIDIA, Huang Win Top Honors in Innovation, Engineering
NVIDIA today was named the world's most innovative company by Fast Company magazine. The accolade comes on the heels of company founder and CEO Jensen Huan...
18/03/2024
NVIDIA Edify Unlocks 3D Generative AI, New Image Controls for Visual Content Providers
NVIDIA Edify, a multimodal architecture for visual generative AI, is entering a ...
18/03/2024
From Atoms to Supercomputers: NVIDIA, Partners Scale Quantum Computing
The latest advances in quantum computing include investigating molecules, deploying giant supercomputers and building the quantum workforce with a new academic ...
18/03/2024
New NVIDIA Storage Partner Validation Program Streamlines Enterprise AI Deployments
A sharp increase in generative AI deployments is driving business innovation for...
18/03/2024
NVIDIA Unveils Digital Blueprint for Building Next-Gen Data Centers
Designing, simulating and bringing up modern data centers is incredibly complex, involving multiple considerations like performance, energy efficiency and scala...