Setting a New Bar in MLPerf

NVIDIA training and inference solutions deliver record-setting performance in MLPerf, the leading industry benchmark for AI performance.

See Our Results

What is MLPerf?

MLPerf is a consortium of AI leaders from academia, research labs, and industry whose mission is to “build fair and useful benchmarks” that provide unbiased evaluations of training and inference performance for hardware, software, and services—all conducted under prescribed conditions. To stay on the cutting edge of industry trends, MLPerf continues to evolve, holding new tests at regular intervals and adding new workloads that represent the state of the art in AI.

Chalmers University is one of the leading research institutions in Sweden, specializing in multiple areas from nanotechnology to climate studies. As we incorporate AI to advance our research endeavors, we find that the MLPerf benchmark provides a transparent apples-to-apples comparison across multiple AI platforms to showcase actual performance in diverse real-world use cases.

— Chalmers University of Technology, Sweden

TSMC is driving the cutting edge of global semiconductor manufacturing, like our latest 5nm node which leads the market in process technology. Innovations like machine learning based lithography and etch modeling dramatically improve our optical proximity correction (OPC) and etch simulation accuracy. To fully realize the potential of machine learning in model training and inference, we are working with the NVIDIA engineering team to port our Maxwell simulation and inverse lithography technology (ILT) engine to GPUs and see very significant speedups. The MLPerf benchmark is an important factor in our decision making.

— Dr. Danping Peng, Director, OPC Department, TSMC, San Jose, CA, USA

Computer vision and imaging are at the core of AI research, driving scientific discovery and readily representing core components of medical care. We've worked closely with NVIDIA to bring innovations like 3DUNet to the healthcare market. Industry-standard MLPerf benchmarks provide relevant performance data to the benefit of IT organizations and developers to get the right solution to accelerate their specific projects and applications.

— Prof. Dr. Klaus Maier-Hein (Head of Medical Image Computing, Deutsches Krebsforschungszentrum (DKFZ, German Cancer Research Center)

As the pre-eminent leader in research and manufacturing, Samsung uses AI to dramatically boost product performance and manufacturing productivity. Productizing these AI advances requires us to have the best computing platform available. The MLPerf benchmark streamlines our selection process by providing us with an open, direct evaluation method to assess uniformly across platforms.

— Samsung Electronics

MLPerf Submission Categories

MLPerf Training v1.0 is the fourth instantiation for training and consists of eight different workloads covering a broad diversity of use cases, including vision, language, recommenders, and reinforcement learning.

MLPerf Inference v1.0 tested seven different use cases across seven different kinds of neural networks. Three of these use cases were for computer vision, one for recommender systems, two for language processing, and one for medical imaging.

Image Classification

Image Classification

Assigns a label from a fixed set of categories to an input image, i.e., applies to computer vision problems. details.

Object Detection (Lightweight)

Object Detection (Lightweight)

Finds instances of real-world objects such as faces, bicycles, and buildings in images or videos and specifies a bounding box around each. details.

Object Detection (Heavyweight)

Object Detection (Heavyweight)

Detects distinct objects of interest appearing in an image and identifies a pixel mask for each. details.

Biomedical Image Segmentation

Biomedical Image Segmentation

Performs volumetric segmentation of dense 3D images for medical use cases. details.

Translation (Recurrent)

Translation (Recurrent)

Translates text from one language to another using a recurrent neural network (RNN). details.

Automatic Speech Recognition (ASR)

Automatic Speech Recognition (ASR)

Recognize and transcribe audio in real time. details.

Natural Language Processing (NLP)

Natural Language Processing (NLP)

Understands text by using the relationship between different words in a block of text. Allows for question answering, sentence paraphrasing, and many other language-related use cases. details.

Recommendation

Recommendation

Delivers personalized results in user-facing services such as social media or e-commerce websites by understanding interactions between users and service items, like products or ads. details.

Reinforcement Learning

Reinforcement Learning

Evaluates different possible actions to maximize reward using the strategy game Go played on a 19x19 grid. details.

NVIDIA’s MLPerf Benchmark Results

  • Training

    Training

  • Inference

    Inference

The NVIDIA A100 Tensor Core GPU and the NVIDIA DGX SuperPOD™ set all 16 training performance records, both in per-chip and at-scale workloads. This breakthrough performance came from the tight integration of hardware, software, and system-level technologies and NVIDIA’s continuous investment in full-stack performance has led to an improvement in throughput over the four MLPerf submissions. The NVIDIA platform excels in both performance and usability, offering a single leadership platform from the data center to the edge and to the cloud.

OVER 20X PERFORMANCE IN 3 YEAR​S OF MLPERF

NVIDIA's Full-Stack Innovation Delivers Continuous Improvements

MLPerf Training Performance Benchmarks

Results normalized for throughput due to higher accuracy requirements on latest round of MLPerf 1.1
MLPerf ID 0.5/0.7/1.1 comparison:
SSD: 0.5-23/0.7-53/1.1-2078 | BERT: 0.7-56/1.1-2083 | ResNet50v1.5: 0.5-17/0.7-55/1.1-2082 | Mask R-CNN: 0.5-22/0.7-48/1.1-2076 | MLPerf name and logo are trademarks. See www.mlperf.org for more information.

NVIDIA SETS ALL 16 RECORDS

For Commercially Available Solutions

The NVIDIA AI platform set all 8 per-accelerator records using NVIDIA A100 GPUs in OEM servers as well as NVIDIA DGX. This demonstrates the strength of the end-to-end NVIDIA hardware and software stack which allows computer makers to deliver record results on MLPerf.

BENCHMARK At Scale Records (min) Per-Accelerator Records (min)
Recommendation (DLRM) 0.63 (DGX SuperPOD) 13.5 (A100)
NLP (BERT) 0.24 (DGX SuperPOD) 155.1 (A100)
Speech Recognition - Recurrent (RNN-T) 2.38 (DGX SuperPOD) 267.0 (A100)
Object Detection - Heavyweight (Mask R-CNN) 3.24 (DGX SuperPOD) 335.1 (A100)
Object Detection - Lightweight (SSD) 0.45 (DGX SuperPOD) 63.8 (A100)
Image Classification (ResNet-50 v1.5) 0.35 (DGX SuperPOD) 220.5 (A100)
Image Segmentation (3D-Unet) 1.26 (Microsoft Azure NDm A100 v4) 187.7 (A100)
Reinforcement Learning (MiniGo) 15.47 (DGX SuperPOD) 1805.2 (A100)

Per Accelerator performance for A100 computed using NVIDIA partner single node time-to-train and multiplying it by number of chips | Per Chip Performance comparisons to others arrived at by comparing performance at the closest similar scale.
Max Scale Records: DLRM: 1.1-2073  |  BERT: 1.1-2083  |  Mask R-CNN: 1.1-2076  |  Resnet50 v1.5: 1.1-2082  |  SSD: 1.1-2078  |  RNN-T: 1.1-2080  |  3D-Unet: 1.1-2012  |  MiniGo: 1.1-2081
Per-Accelerator Records: DLRM: 1.1-2084  |  BERT: 1.1-2053  |  Mask R-CNN: 1.1-2031  |  Resnet50 v1.5: 1.1-2050  |  SSD: 1.1-2050  |  RNN-T: 1.1-2052  |  3D-Unet: 1.1-2051  |  MiniGo: 1.1-2031
MLPerf name and logo are trademarks. See www.mlperf.org for more information.

NVIDIA achieved top performance results in all scenarios (data center server and offline, as well as edge single-stream, multi-stream, and offline). In addition, we delivered the best per-accelerator performance among all products tested across all benchmark tests. These results are a testament, not only to NVIDIA's inference performance leadership, but to the versatility of our inference platform.

Offline scenario for data center and edge (Single GPU)

  NVIDIA A100 (x86 CPU)
(Inferences/Second)		 NVIDIA A100 (Arm CPU)
(Inferences/Second)		 NVIDIA A30
(Inferences/Second)		 NVIDIA® Jetson Xavier™
(Max Inferences/Query)
DLRM
(Recommender)		 302,680 274,433 135,450 N/A*
BERT
(Natural Language Processing)		 3,538 3,151 1,673 97
ResNet-50 v1.5
(Image Classification)		 39,190 36,436 18,647 2,039
ResNet-34
(Large Single-Shot Detector)		 981 901 474 55
RNN-T
(Speech Recognition)		 13,365 12,640 6,574 416
3D U-Net
(Medical Imaging)		 61 57 30 3

*DLRM not part of Edge category suite.

1) MLPerf v1.1 Inference results for data center server form factors and offline scenario retrieved from www.mlperf.org on September 22, 2021 from entries Inf-1.1-047, Inf-1.1-43, Inf-1.1-023-026, Per-processor performance is calculated by dividing the primary metric of total performance by the number of accelerators reported.

2) MLPerf v1.1 Inference results for edge form factors on offline scenario retrieved from www.mlperf.org on September, 2021 from entries 1.1-117 Per-processor performance is calculated by dividing the primary metric of total performance by the number of accelerators reported.

The Technology Behind the Results

The complexity of AI demands a tight integration between all aspects of the platform. As demonstrated in MLPerf's benchmarks, the NVIDIA AI platform delivers leadership performance with the world's most advanced GPU, powerful and scalable interconnect technologies, and cutting-edge software—an end-to-end solution that can be deployed in the data center, in the cloud, or at the edge with amazing results.

Pre-trained models and Optimized Software from NVIDIA NGC

Optimized Software that Accelerates AI Workflows

An essential component of NVIDIA’s platform and MLPerf training and inference results, NGC™ is a hub for GPU-optimized AI, high-performance computing (HPC), and data analytics software that simplifies and accelerates end-to-end workflows. With over 150 enterprise-grade containers, including workloads for conversational AI and recommender systems, over 100 models, and industry-specific SDKs that can be deployed on-premises, in the cloud, or at the edge, NGC enables data scientists, researchers, and developers to build best-in-class solutions, gather insights, and deliver business value faster than ever before.

VISIT NGC

Leadership-Class AI Infrastructure

Achieving world-leading results across training and inference requires infrastructure that’s purpose-built for the world’s most complex AI challenges. The NVIDIA AI platform delivered using the power of the NVIDIA A100 Tensor Core GPU, the NVIDIA A30 Tensor Core GPU, the NVIDIA A2 Tensor Core GPU, and the scalability and flexibility of NVIDIA interconnect technologies—NVIDIA NVLink®, NVIDIA NVSwitch™, and the NVIDIA ConnectX®-6 VPI. These are at the heart of the NVIDIA DGX™ A100, the engine behind our benchmark performance.

NVIDIA DGX™ systems offer the scalability, rapid deployment, and incredible compute power that enables every enterprise to build leadership-class AI infrastructure.

LEARN MORE
NVIDIA Tensor Core GPUs

Learn more about our data center training and inference product performance.

View Performance
Image Classification

Image Classification

MLPerf Training and Inference uses ResNet v1.5 with the ImageNet dataset.
MLPerf Inference used ResNet-50 v. 1.5 and MobileNet-v1 with the ImageNet dataset.
Object Detection (Lightweight)

Object Detection (Lightweight)

MLPerf Training uses Single-Shot Detector (SSD) on the COCO 2017 dataset.
MLPerf Inference uses SSD-MobileNet-v1 (low-res, 0.09 MP), with the COCO 2017 dataset.
Object Detection (Heavyweight)

Object Detection (Heavyweight)

MLPerf Training uses Mask R-CNN on the COCO 2014 dataset.
MLPerf Inference uses SSD-ResNet-34 (high-res, 1.44 MP) with the COCO 2014 dataset.
Translation (Recurrent)

Translation (Recurrent)

MLPerf used Google’s Machine Translation System (GNMT) on the WMT English-German dataset.
MLPerf Inference used the Google Neural Machine Translation (GNMT) RNN with the WMT16 dataset.
Translation (Non-recurrent)

Translation (Non-recurrent)

MLPerf used Transformer on the WMT English-German dataset.
MLPerf Inference measured speech recognition using the RNN-T network with the LibriSpeech dataset.
Natural Language Processing (NLP)

Natural Language Processing (NLP)

MLPerf Training uses Bidirectional Encoder Representations from Transformers (BERT) on the Wikipedia 2020/01/01 dataset.
MLPerf Inference used BERT with the SQuAD v.1.1 dataset.
Recommendation

Recommendation

MLPerf Training and Inference uses the Deep Learning Recommendation Model (DLRM) on Criteo 1 TB dataset.
MLPerf Inference used the DLRM network with the Criteo Terabyte dataset.
Reinforcement Learning

Reinforcement Learning

MLPerf Training uses Mini Go benchmark based gameplay.
Automatic Speech Recognition (ASR

Automatic Speech Recognition (ASR)

MLPerf Training and Inference uses RNN-T on the Librispeech dataset.
Biomedical Image Segmentation

Biomedical Image Segmentation

MLPerf Training and Inference uses UNet-3D with KiTS19 and BraTS 2019 dataset respectively.

Products

Technologies

Software

Resources