![](//www.4rfv.com/images/no_image.jpg)
It's official: NVIDIA delivered the world's fastest platform in industry-standard tests for inference on generative AI.
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
25/07/2024
Hey, you. You're finally awake.
It's the summer of Elder Scrolls - whe...
24/07/2024
AI is set to transform the workforce - and the Georgia Institute of Technology's new AI Makerspace is helping tens of thousands of students get ahead of the...
24/07/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and showcases new hardware, softwa...
23/07/2024
Generative AI applications have little, or sometimes negative, value without acc...
23/07/2024
Businesses seeking to harness the power of AI need customized models tailored to their specific industry needs.
NVIDIA AI Foundry is a service that enables ent...
22/07/2024
AI and accelerated computing - twin engines NVIDIA continuously improves - are d...
22/07/2024
Team NVIDIA has triumphed at the Amazon KDD Cup 2024, securing first place Friday across all five competition tracks.
The team - consisting of NVIDIANs Ahmet E...
19/07/2024
Research published earlier this month in the science journal Nature used NVIDIA-powered supercomputers to validate a pathway toward the commercialization of qua...
19/07/2024
Research published earlier this month in the science journal Nature used NVIDIA-powered supercomputers to validate a pathway toward the commercialization of qua...
19/07/2024
AI has seen unprecedented growth - spurring the need for new training and educat...
19/07/2024
Research published earlier this month in the science journal Nature used NVIDIA-powered supercomputers to validate a pathway toward the commercialization of qua...
18/07/2024
Mistral AI and NVIDIA today released a new state-of-the-art language model, Mist...
18/07/2024
It's time for a sweet treat - the GeForce NOW Summer Sale offers high-perfor...
17/07/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and showcases new hardware, softwa...
16/07/2024
Editor's note: This post is part of our In the NVIDIA Studio series, which c...
15/07/2024
NVIDIA founder and CEO Jensen Huang and Meta founder and CEO Mark Zuckerberg wil...
12/07/2024
NVIDIA is taking an array of advancements in rendering, simulation and generativ...
11/07/2024
Unlock new experiences every GFN Thursday. Whether post-apocalyptic survival adventures, narrative-driven games or vast, open worlds, GeForce NOW always has som...
11/07/2024
Enhancing Japan's AI sovereignty and strengthening its research and development capabilities, Japan's National Institute of Advanced Industrial Science ...
10/07/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible and showcases new hardware, softwar...
10/07/2024
Improved cancer diagnostics - and improved patient outcomes - could be among the...
09/07/2024
Sphere, a new kind of entertainment medium in Las Vegas, is joining the ranks of legendary circular performance spaces such as the Roman Colosseum and Shakespea...
08/07/2024
Artificial intelligence is transforming the transportation industry, helping dri...
04/07/2024
GeForce NOW is bringing 22 new games to members this month.
Dive into the four titles available to stream on the cloud gaming service this week to stay cool an...
03/07/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and showcases new hardware, softwa...
28/06/2024
On a weekday afternoon, Ashwini Ashtankar sat on the bank of the Doodhpathri River, in a valley nestled in the Himalayas. Taking a deep breath, she noticed that...
27/06/2024
Editor's note: This post is part of Into the Omniverse, a series focused on ...
27/06/2024
Get ready to feel some chills, even amid the summer heat. Capcom's award-winning Resident Evil Village brings a touch of horror to the cloud this GFN Thursd...
26/06/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and showcases new hardware, softwa...
26/06/2024
Roblox is a colorful online platform that aims to reimagine the way that people ...
25/06/2024
Generative AI has revolutionized software development with prompt-based code generation - protein design is next.
EvolutionaryScale today announced the release...
24/06/2024
Multi-die chips, known as three-dimensional integrated circuits, or 3D-ICs, represent a revolutionary step in semiconductor design. The chips are vertically sta...
20/06/2024
Sit back and settle in for some epic storytelling. Tell Me Why and As Dusk Falls - award-winning, narrative-driven games from Xbox Studios - add to the 1,900+ g...
19/06/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible and showcases new hardware, softwar...
18/06/2024
The electric grid and the utilities managing it have an important role to play in the next industrial revolution that's being driven by AI and accelerated c...
17/06/2024
NVIDIA contributed the largest ever indoor synthetic dataset to the Computer Vision and Pattern Recognition (CVPR) conference's annual AI City Challenge - h...
17/06/2024
Making moves to accelerate self-driving car development, NVIDIA was today named an Autonomous Grand Challenge winner at the Computer Vision and Pattern Recognit...
17/06/2024
NVIDIA researchers are at the forefront of the rapidly advancing field of visual...
15/06/2024
NVIDIA founder and CEO Jensen Huang on Friday encouraged Caltech graduates to pu...
14/06/2024
When she was five years old, Veronica Miller (n e Teklai) and her family left their homeland of Eritrea, in the Horn of Africa, to escape an ongoing war with Et...
14/06/2024
NVIDIA today announced Nemotron-4 340B, a family of open models that developers ...
13/06/2024
Set sail for adventure, pirates. Sea of Thieves makes waves in the cloud this week. It's an adventure-filled GFN Thursday with four new games joining the Ge...
12/06/2024
Accelerated computing is transforming data processing and analytics for enterpri...
12/06/2024
The full-stack NVIDIA accelerated computing platform has once again demonstrated...
12/06/2024
Let's talk about NeRFs - no, not the neon-colored foam dart blasters, but ne...
12/06/2024
Editor's note: This post is part of the AI Decoded series, which demystifies AI by making the technology more accessible, and showcases new hardware, softwa...
07/06/2024
Across industries, AI is supercharging innovation with machine-powered computation. In finance, bankers are using AI to detect fraud more quickly and keep accou...
06/06/2024
Capcom's latest entry in the iconic Street Fighter series, Street Fighter 6, punches its way into the cloud this GFN Thursday. The game, along with Ubisoft&...
05/06/2024
India's AI market is expected to be massive. Yotta Data Services is setting its sights on supercharging it. In this episode of NVIDIA's AI Podcast, Suni...
05/06/2024
NVIDIA launched NVIDIA Studio at COMPUTEX in 2019. Five years and more than 500 ...