PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark].

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

Project Site

pytorch.org

Source Repository

github.com

Test Created

16 November 2023

Last Updated

26 March 2024

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

4 Minutes, 56 Seconds

Average Run Time

12 Minutes, 40 Seconds

Test Dependencies

Python

Accolades

10k+ Downloads

Supported Platforms

Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page ViewsOpenBenchmarking.orgEventsPyTorch Popularity Statisticspts/pytorch2023.112023.122024.012024.022024.032024.042024.052024.062024.072024.082024.092024.102024.114K8K12K16K20K
* 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 24 November 2024.
NVIDIA CUDA GPU5.1%CPU94.9%Device Option PopularityOpenBenchmarking.org
1616.2%120.5%25617.4%6420.2%51216.4%329.2%Batch Size Option PopularityOpenBenchmarking.org
ResNet-5054.4%ResNet-15231.7%Efficientnet_v2_l13.9%Model Option PopularityOpenBenchmarking.org

Revision History

pts/pytorch-1.1.0   [View Source]   Tue, 26 Mar 2024 15:55:49 GMT
Update to PyTorch 2.2.1 to allow working on Ubuntu 24.04.

pts/pytorch-1.0.1   [View Source]   Sat, 06 Jan 2024 15:54:34 GMT
Add NVIDIA CUDA GPU option.

pts/pytorch-1.0.0   [View Source]   Thu, 16 Nov 2023 15:22:58 GMT
Initial commit.

Suites Using This Test

Machine Learning

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

PyTorch 2.2.1

Device: CPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.org metrics for this test profile configuration based on 389 public results since 26 March 2024 with the latest data as of 10 November 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
batches/sec (Average)
AMD Ryzen 7 9800X3D 8-Core
100th
6
78.7 +/- 0.2
AMD Ryzen 7 9700X 8-Core
99th
10
64.6 +/- 1.9
INTEL XEON PLATINUM 8592
97th
4
63.1
AMD Ryzen 9 9950X 16-Core
94th
16
58.4 +/- 1.5
Intel Core Ultra 9 285K
93rd
8
57.7 +/- 2.9
Intel Xeon Platinum 8490H
91st
4
56.0
AMD Ryzen 5 9600X 6-Core
88th
10
55.5 +/- 1.0
AMD Ryzen 9 9900X 12-Core
85th
7
52.5 +/- 0.5
AMD Ryzen 9 7950X 16-Core
82nd
11
48.9 +/- 2.0
AMD Ryzen Threadripper 7980X 64-Cores
80th
4
47.9 +/- 0.3
AMD EPYC 9655 96-Core
79th
3
47.8 +/- 1.5
AMD Ryzen 7 7700X 8-Core
79th
4
47.2
AMD EPYC 4584PX 16-Core
77th
4
46.2
AMD Ryzen 7 7700 8-Core
76th
4
46.0
Mid-Tier
75th
< 46.0
AMD Ryzen 7 7800X3D 8-Core
75th
5
45.3
AMD EPYC 4364P 8-Core
73rd
4
45.1 +/- 0.1
Intel Core Ultra 5 245K
72nd
5
44.6 +/- 0.1
AMD EPYC 4344P 8-Core
71st
3
43.9
AMD EPYC 4564P 16-Core
70th
5
43.9 +/- 1.1
AMD EPYC 9755 128-Core
69th
8
43.5 +/- 0.2
AMD Ryzen 9 7900X3D 12-Core
67th
5
43.3
AMD Ryzen 9 7900 12-Core
65th
5
42.6
Intel Core i9-13900K
62nd
3
42.3
Intel Core i9-14900K
61st
9
42.2 +/- 1.3
2 x INTEL XEON PLATINUM 8592
61st
6
41.8 +/- 1.0
AMD EPYC 4484PX 12-Core
60th
4
41.3
AMD EPYC 4464P 12-Core
59th
3
39.5
AMD Ryzen 9 7950X3D 16-Core
58th
5
38.4
AMD Ryzen 5 7600 6-Core
54th
4
35.5
AMD EPYC 4244P 6-Core
53rd
3
34.9
AMD EPYC 9684X 96-Core
52nd
3
34.7
AMD Ryzen 9 7900X 12-Core
51st
5
34.3
Median
50th
34.3
AMD EPYC 8534P 64-Core
49th
5
33.6 +/- 0.2
AMD Ryzen 7 5800X3D 8-Core
49th
5
33.5
AMD Ryzen 9 5950X 16-Core
45th
5
31.9
AMD Ryzen Threadripper PRO 5965WX 24-Cores
43rd
4
31.7 +/- 0.1
Intel Xeon E E-2488
43rd
9
31.6 +/- 4.6
AMD Ryzen 7 8700G
42nd
5
31.6
AMD Ryzen 5 8400F 6-Core
41st
4
31.1
AMD Ryzen 7 5800X 8-Core
40th
5
31.1
Intel Xeon 6780E
39th
4
30.7
AMD Ryzen 5 8600G
38th
5
30.4 +/- 4.1
Intel Core i5-13600K
37th
3
30.1
AMD Ryzen Threadripper 3970X 32-Core
36th
3
30.0 +/- 0.2
AMD Ryzen 9 5900X 12-Core
36th
6
29.7
Intel Core i5-14600K
33rd
5
29.1
AMD Ryzen AI 9 HX 370
31st
3
28.1 +/- 2.2
AMD EPYC 9754 128-Core
31st
3
27.7
Intel Xeon 6766E
30th
4
26.9
AMD Ryzen 5 7600X 6-Core
28th
4
25.1
Low-Tier
25th
< 23.2
AMD EPYC 4124P 4-Core
25th
4
23.1 +/- 0.2
Intel Core i5-14500
25th
4
22.7
AMD Ryzen 5 5600X 6-Core
23rd
4
21.6
2 x AMD EPYC 9684X 96-Core
19th
8
20.8 +/- 1.4
Intel Core i3-14100
19th
5
20.4
AmpereOne 192-Core
18th
3
20.1
AMD Ryzen 5 5500
17th
5
19.4
Intel Core i5-13400
16th
3
18.9
AMD Ryzen 9 5900HX
15th
5
18.7 +/- 1.6
Intel Core i5-14400F
14th
5
18.2
2 x Intel Xeon 6980P
12th
3
17.0 +/- 0.5
Intel Core Ultra 7 155H
9th
5
15.2 +/- 0.9
AMD Ryzen 7 PRO 5850U
9th
4
15.2
Intel Core i7-1280P
8th
5
15.1 +/- 0.7
AMD Ryzen 5 5500U
5th
4
12.8 +/- 0.1
Intel Core i7-1065G7
3rd
4
8.6 +/- 0.1
OpenBenchmarking.orgDistribution Of Public Results - Device: CPU - Batch Size: 64 - Model: ResNet-50389 Results Range From 6 To 79 batches/sec613202734414855626976839020406080100

Based on OpenBenchmarking.org data, the selected test / test configuration (PyTorch 2.2.1 - Device: CPU - Batch Size: 64 - Model: ResNet-50) has an average run-time of 6 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 BenchmarkDevice: CPU - Batch Size: 64 - Model: ResNet-50Run-Time510152025Min: 1 / Avg: 5.24 / Max: 21

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.9%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsDevice: CPU - Batch Size: 64 - Model: ResNet-50Deviation3691215Min: 0 / Avg: 0.94 / Max: 8

Does It Scale Well With Increasing Cores?

No, based on the automated analysis of the collected public benchmark data, this test / test settings does not generally scale well with increasing CPU core counts. Data based on publicly available results for this test / test settings, separated by vendor, result divided by the reference CPU clock speed, grouped by matching physical CPU core count, and normalized against the smallest core count tested from each vendor for each CPU having a sufficient number of test samples and statistically significant data.

AMDIntelOpenBenchmarking.orgRelative Core Scaling To BasePyTorch CPU Core ScalingDevice: CPU - Batch Size: 64 - Model: ResNet-5068121416240.48750.9751.46251.952.4375

Notable Instruction Set Usage

Notable instruction set extensions supported by this test, based on an automatic analysis by the Phoronix Test Suite / OpenBenchmarking.org analytics engine.

Instruction Set
Support
Instructions Detected
SSE2 (SSE2)
Used by default on supported hardware.
 
MOVDQA MOVDQU PUNPCKLQDQ PADDQ CVTSS2SD UCOMISD CVTSD2SS MOVD CVTSI2SD ADDSD MULSD XORPD DIVSD SUBSD CVTTSD2SI COMISD MOVAPD SHUFPD ANDPD UNPCKLPD MULPD UNPCKHPD SUBPD ADDPD PUNPCKHQDQ PSUBQ PSHUFD PSRLDQ ORPD SQRTSD ANDNPD CMPNLESD MINSD MAXSD MOVMSKPD MOVUPD CMPLTSD DIVPD
Last automated analysis: 6 April 2024

This test profile binary relies on the shared libraries libm.so.6, libexpat.so.1, libz.so.1, libc.so.6.

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, AmpereOne 192-Core

Recent Test Results

OpenBenchmarking.org Results Compare

1 System - 323 Benchmark Results

AMD Ryzen Threadripper 7960X 24-Cores - Gigabyte TRX50 AERO D - AMD Device 14a4

Ubuntu 24.04 - 6.8.0-48-generic - GNOME Shell 46.0

2 Systems - 541 Benchmark Results

ARMv8 Cortex-A76 - Raspberry Pi 5 Model B Rev 1.0 - Broadcom BCM2712

Ubuntu 24.04 - 6.8.0-1012-raspi - KDE Plasma 5.27.11

2 Systems - 525 Benchmark Results

ARMv8 Cortex-A76 - Raspberry Pi 5 Model B Rev 1.0 - Broadcom BCM2712

Ubuntu 24.04 - 6.8.0-1012-raspi - KDE Plasma 5.27.11

1 System - 342 Benchmark Results

Intel Core i9-12900K - ASUS PRIME Z790-V AX - Intel Raptor Lake-S PCH

Ubuntu 24.04 - 6.8.0-47-generic - GNOME Shell 46.0

1 System - 18 Benchmark Results

AMD Ryzen Threadripper 3960X 24-Core - ASUS ROG ZENITH II EXTREME ALPHA - AMD Starship

Fedora Linux 40 - 6.11.3-200.fc40.x86_64 - GNOME Shell 46.5

1 System - 495 Benchmark Results

3 Systems - 242 Benchmark Results

AMD Ryzen 7 7840HS - NB05 TUXEDO Pulse 14 Gen3 R14FA1 - AMD Device 14e8

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

Most Popular Test Results

OpenBenchmarking.org Results Compare

3 Systems - 413 Benchmark Results

AMD Ryzen 7 7840HS - Framework Laptop 16 - AMD Device 14e8

Ubuntu 24.04 - 6.10.0-061000rc4daily20240621-generic - GNOME Shell 46.0

18 Systems - 442 Benchmark Results

AMD Ryzen 9 7900X 12-Core - ASUS ROG STRIX X670E-E GAMING WIFI - AMD Device 14d8

Ubuntu 24.04 - 6.10.0-061000rc2-generic - GNOME Shell 46.0

3 Systems - 18 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

5 Systems - 464 Benchmark Results

AMD EPYC 4564P 16-Core - Supermicro AS-3015A-I H13SAE-MF v1.00 - AMD Device 14d8

Ubuntu 24.04 - 6.8.0-11-generic - GNOME Shell 45.3

2 Systems - 57 Benchmark Results

AMD Ryzen Threadripper PRO 5965WX 24-Cores - ASUS Pro WS WRX80E-SAGE SE WIFI - AMD Starship

Ubuntu 23.10 - 6.5.0-26-generic - GNOME Shell 45.0

2 Systems - 42 Benchmark Results

AMD Ryzen 9 5900HX - ASUS ROG Strix G513QY_G513QY G513QY v1.0 - AMD Renoir

Ubuntu 22.10 - 5.19.0-46-generic - GNOME Shell 43.0

2 Systems - 108 Benchmark Results

Intel Core i7-1065G7 - Dell 06CDVY - Intel Ice Lake-LP DRAM

Ubuntu 23.10 - 6.7.0-060700rc5-generic - GNOME Shell 45.1

3 Systems - 55 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

2 Systems - 42 Benchmark Results

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

Tuxedo 22.04 - 6.5.0-10027-tuxedo - KDE Plasma 5.27.10

2 Systems - 49 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

4 Systems - 18 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

2 Systems - 12 Benchmark Results

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

Ubuntu 23.10 - 6.6.0-rc5-phx-patched - GNOME Shell 45.0

22 Systems - 263 Benchmark Results

2 x Intel Xeon Platinum 8490H - Quanta Cloud QuantaGrid D54Q-2U S6Q-MB-MPS - Intel Device 1bce

Ubuntu 24.04 - 6.9.0-060900rc3-generic - GCC 13.2.0

4 Systems - 18 Benchmark Results

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

Ubuntu 23.10 - 6.5.0-26-generic - GNOME Shell 45.2
Find More Test Results