Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/).

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark deepsparse.

Project Site

neuralmagic.com

Source Repository

github.com

Test Created

13 October 2022

Last Updated

15 March 2024

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

29 Minutes, 23 Seconds

Average Run Time

3 Minutes, 8 Seconds

Test Dependencies

Python

Accolades

30k+ Downloads + Recently Updated Test Profile

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page ViewsOpenBenchmarking.orgEventsNeural Magic DeepSparse Popularity Statisticspts/deepsparse2022.102022.112022.122023.012023.022023.032023.042023.052023.062023.072023.082023.092023.102023.112023.122024.012024.022024.035K10K15K20K25K
* Uploading of benchmark result data to OpenBenchmarking.org is always optional (opt-in) via the Phoronix Test Suite for users wishing to share their results publicly.
** Data based on those opting to upload their test results to OpenBenchmarking.org and users enabling the opt-in anonymous statistics reporting while running benchmarks from an Internet-connected platform.
Data updated weekly as of 17 March 2024.
ResNet-50, Sparse INT88.7%NLP Text Classification, BERT base uncased SST2, Sparse INT88.3%NLP Token Classification, BERT base uncased conll20037.9%BERT-Large, NLP Question Answering8.6%CV Classification, ResNet-50 ImageNet8.5%NLP Text Classification, DistilBERT mnli8.0%CV Detection, YOLOv5s COCO, Sparse INT88.0%NLP Document Classification, oBERT base uncased on IMDB8.3%BERT-Large, NLP Question Answering, Sparse INT88.6%CV Detection, YOLOv5s COCO8.2%ResNet-50, Baseline8.3%CV Segmentation, 90% Pruned YOLACT Pruned8.6%Model Option PopularityOpenBenchmarking.org
Asynchronous Multi-Stream65.2%Synchronous Single-Stream34.8%Scenario Option PopularityOpenBenchmarking.org

Revision History

pts/deepsparse-1.7.0   [View Source]   Fri, 15 Mar 2024 12:35:17 GMT
Update against DeepSparse 1.7 upstream, add Llama 2 chat test.

pts/deepsparse-1.6.0   [View Source]   Mon, 11 Dec 2023 16:59:10 GMT
Update against deepsparse 1.6 upstream.

pts/deepsparse-1.5.2   [View Source]   Wed, 26 Jul 2023 15:52:28 GMT
Update against 1.5.2 point release, add more models.

pts/deepsparse-1.5.0   [View Source]   Wed, 07 Jun 2023 07:51:58 GMT
Update against Deepsparse 1.5 upstream.

pts/deepsparse-1.3.2   [View Source]   Sun, 22 Jan 2023 19:05:03 GMT
Update against DeepSparse 1.3.2 upstream.

pts/deepsparse-1.0.1   [View Source]   Thu, 13 Oct 2022 13:47:39 GMT
Initial commit of DeepSparse benchmark.

Suites Using This Test

Machine Learning

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

Neural Magic DeepSparse 1.7

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream

OpenBenchmarking.org metrics for this test profile configuration based on 53 public results since 15 March 2024 with the latest data as of 17 March 2024.

Below is an overview of the generalized performance for components where there is sufficient statistically significant data based upon user-uploaded results. It is important to keep in mind particularly in the Linux/open-source space there can be vastly different OS configurations, with this overview intended to offer just general guidance as to the performance expectations.

Component
Percentile Rank
# Compatible Public Results
ms/batch (Average)
Mid-Tier
75th
> 39
Median
50th
63
38th
5
105 +/- 1
Low-Tier
25th
> 176
16th
4
179 +/- 1
2nd
3
239 +/- 12
OpenBenchmarking.orgDistribution Of Public Results - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream39 Results Range From 20 To 253 ms/batch20355065809511012514015517018520021523024526048121620

Based on OpenBenchmarking.org data, the selected test / test configuration (Neural Magic DeepSparse 1.7 - Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream) has an average run-time of 3 minutes. By default this test profile is set to run at least 3 times but may increase if the standard deviation exceeds pre-defined defaults or other calculations deem additional runs necessary for greater statistical accuracy of the result.

OpenBenchmarking.orgMinutesTime Required To Complete BenchmarkModel: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamRun-Time246810Min: 2 / Avg: 2 / Max: 2

Tested CPU Architectures

This benchmark has been successfully tested on the below mentioned architectures. The CPU architectures listed is where successful OpenBenchmarking.org result uploads occurred, namely for helping to determine if a given test is compatible with various alternative CPU architectures.

CPU Architecture
Kernel Identifier
Verified On
Intel / AMD x86 64-bit
x86_64
(Many Processors)
ARMv8 64-bit
aarch64
ARMv8 Neoverse-N1 128-Core

Recent Test Results

OpenBenchmarking.org Results Compare

2 Systems - 57 Benchmark Results

AMD Ryzen 7 3800XT 8-Core - MSI X370 XPOWER GAMING TITANIUM - AMD Starship

Debian 12 - 6.1.0-18-amd64 - X Server 1.20.11

2 Systems - 55 Benchmark Results

AMD Ryzen 7 7840U - PHX Swift SFE16-43 Ray_PEU - AMD Device 14e8

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

2 Systems - 63 Benchmark Results

Intel Core i9-10980XE - ASRock X299 Steel Legend - Intel Sky Lake-E DMI3 Registers

Ubuntu 22.04 - 6.5.0-18-generic - GNOME Shell 42.2

4 Systems - 54 Benchmark Results

2 x AMD EPYC 9684X 96-Core - AMD Titanite_4G - AMD Device 14a4

Ubuntu 23.10 - 6.5.0-25-generic - GCC 13.2.0

3 Systems - 61 Benchmark Results

AMD Ryzen Threadripper 3970X 32-Core - ASUS ROG ZENITH II EXTREME - AMD Starship

Ubuntu 22.04 - 6.5.0-21-generic - GNOME Shell 42.2

3 Systems - 114 Benchmark Results

AMD Ryzen 5 5500U - NB01 TUXEDO Aura 15 Gen2 NL5xNU - AMD Renoir

Tuxedo 22.04 - 6.0.0-1010-oem - KDE Plasma 5.26.5

4 Systems - 52 Benchmark Results

AMD Ryzen Threadripper 7980X 64-Cores - System76 Thelio Major - AMD Device 14a4

Ubuntu 23.10 - 6.5.0-25-generic - GNOME Shell 45.2

5 Systems - 44 Benchmark Results

Intel Core i9-14900K - ASUS PRIME Z790-P WIFI - Intel Device 7a27

Ubuntu 23.10 - 6.8.0-phx - GNOME Shell 45.1

3 Systems - 44 Benchmark Results

Intel Core i7-1280P - MSI Prestige 14Evo A12M MS-14C6 - Intel Alder Lake PCH

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

5 Systems - 44 Benchmark Results

AMD Ryzen 9 7950X 16-Core - ASUS ROG STRIX X670E-E GAMING WIFI - AMD Device 14d8

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

1 System - 341 Benchmark Results

AMD Ryzen 9 7950X 16-Core - ASUS ProArt X670E-CREATOR WIFI - AMD Device 14d8

Pop 22.04 - 6.6.10-76060610-generic - GNOME Shell 42.5

4 Systems - 125 Benchmark Results

2 x INTEL XEON PLATINUM 8592+ - Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS - Intel Device 1bce

Fedora Linux 39 - 6.7.6-200.fc39.x86_64 - GCC 13.2.1 20231205

3 Systems - 125 Benchmark Results

2 x INTEL XEON PLATINUM 8592+ - Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS - Intel Device 1bce

Fedora Linux 39 - 6.7.6-200.fc39.x86_64 - GCC 13.2.1 20231205

1 System - 125 Benchmark Results

2 x INTEL XEON PLATINUM 8592+ - Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS - Intel Device 1bce

Ubuntu 23.10 - 6.5.0-17-generic - GCC 13.2.0

1 System - 304 Benchmark Results

AMD A4-5300 APU - ASRock FM2A88M-HD+ R3.0 - AMD 15h

Ubuntu 20.04 - 5.15.0-89-generic - GNOME Shell 3.36.9

Most Popular Test Results

OpenBenchmarking.org Results Compare

5 Systems - 44 Benchmark Results

Intel Core i9-14900K - ASUS PRIME Z790-P WIFI - Intel Device 7a27

Ubuntu 23.10 - 6.8.0-phx - GNOME Shell 45.1

4 Systems - 52 Benchmark Results

AMD Ryzen Threadripper 7980X 64-Cores - System76 Thelio Major - AMD Device 14a4

Ubuntu 23.10 - 6.5.0-25-generic - GNOME Shell 45.2

5 Systems - 44 Benchmark Results

AMD Ryzen 9 7950X 16-Core - ASUS ROG STRIX X670E-E GAMING WIFI - AMD Device 14d8

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

3 Systems - 44 Benchmark Results

Intel Core i7-1280P - MSI Prestige 14Evo A12M MS-14C6 - Intel Alder Lake PCH

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

4 Systems - 54 Benchmark Results

2 x AMD EPYC 9684X 96-Core - AMD Titanite_4G - AMD Device 14a4

Ubuntu 23.10 - 6.5.0-25-generic - GCC 13.2.0

3 Systems - 114 Benchmark Results

AMD Ryzen 5 5500U - NB01 TUXEDO Aura 15 Gen2 NL5xNU - AMD Renoir

Tuxedo 22.04 - 6.0.0-1010-oem - KDE Plasma 5.26.5

2 Systems - 55 Benchmark Results

AMD Ryzen 7 7840U - PHX Swift SFE16-43 Ray_PEU - AMD Device 14e8

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

2 Systems - 63 Benchmark Results

Intel Core i9-10980XE - ASRock X299 Steel Legend - Intel Sky Lake-E DMI3 Registers

Ubuntu 22.04 - 6.5.0-18-generic - GNOME Shell 42.2

3 Systems - 61 Benchmark Results

AMD Ryzen Threadripper 3970X 32-Core - ASUS ROG ZENITH II EXTREME - AMD Starship

Ubuntu 22.04 - 6.5.0-21-generic - GNOME Shell 42.2

2 Systems - 55 Benchmark Results

AMD Ryzen 7 7840U - PHX Swift SFE16-43 Ray_PEU - AMD Device 14e8

Ubuntu 23.10 - 6.7.0-060700-generic - GNOME Shell 45.2

2 Systems - 57 Benchmark Results

AMD Ryzen 7 3800XT 8-Core - MSI X370 XPOWER GAMING TITANIUM - AMD Starship

Debian 12 - 6.1.0-18-amd64 - X Server 1.20.11

2 Systems - 57 Benchmark Results

AMD Ryzen 7 3800XT 8-Core - MSI X370 XPOWER GAMING TITANIUM - AMD Starship

Debian 12 - 6.1.0-18-amd64 - X Server 1.20.11

2 Systems - 63 Benchmark Results

Intel Core i9-10980XE - ASRock X299 Steel Legend - Intel Sky Lake-E DMI3 Registers

Ubuntu 22.04 - 6.5.0-18-generic - GNOME Shell 42.2

Find More Test Results