AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library.

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

Project Site

ai-benchmark.com

Test Created

8 July 2020

Last Updated

11 November 2020

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

2 Minutes, 2 Seconds

Average Run Time

21 Minutes, 23 Seconds

Test Dependencies

Python

Accolades

10k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsAI Benchmark Alpha Popularity Statisticspts/ai-benchmark2020.072020.082020.092020.102020.112020.122021.012021.022021.032021.042021.052021.062021.072021.082021.092021.107001400210028003500
* 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.
*** Test profile page view reporting began March 2021.
Data current as of 22 October 2021.

Revision History

pts/ai-benchmark-1.0.1   [View Source]   Wed, 11 Nov 2020 18:43:35 GMT
Fix for macOS support.

pts/ai-benchmark-1.0.0   [View Source]   Wed, 08 Jul 2020 14:25:53 GMT
Initial commit of AI Benchmark Alpha.

Suites Using This Test

Machine Learning

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

AI Benchmark Alpha 0.1.2

Device AI Score

OpenBenchmarking.org metrics for this test profile configuration based on 1,201 public results since 8 July 2020 with the latest data as of 20 October 2021.

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
Score (Average)
100th
4
4304 +/- 8
98th
4
3616 +/- 148
94th
24
3517 +/- 192
91st
16
3433 +/- 179
91st
3
3429 +/- 130
90th
17
3391 +/- 206
90th
6
3388 +/- 196
90th
16
3380 +/- 100
89th
9
3327 +/- 119
88th
8
3294 +/- 194
87th
8
3226 +/- 141
87th
8
3222 +/- 245
86th
6
3193 +/- 359
85th
9
3141 +/- 155
82nd
10
3054 +/- 155
80th
11
3019 +/- 226
79th
16
2986 +/- 151
78th
10
2921 +/- 400
76th
31
2901 +/- 144
Mid-Tier
75th
< 2865
75th
5
2830 +/- 143
75th
15
2811 +/- 87
74th
8
2766 +/- 129
72nd
3
2726 +/- 7
70th
3
2676 +/- 326
69th
7
2655 +/- 149
68th
5
2617 +/- 155
66th
8
2579 +/- 10
65th
6
2566 +/- 124
65th
10
2553 +/- 106
64th
13
2496 +/- 86
63rd
8
2473 +/- 14
60th
9
2389 +/- 11
59th
23
2358 +/- 142
59th
12
2341 +/- 112
Median
50th
2326
48th
14
2303 +/- 12
46th
13
2227 +/- 12
44th
10
2207 +/- 77
42nd
4
2144 +/- 12
42nd
5
2143 +/- 272
42nd
3
2137 +/- 10
41st
17
2129 +/- 90
41st
4
2084 +/- 4
41st
3
2083 +/- 56
40th
10
2004 +/- 52
27th
3
1856 +/- 113
27th
4
1849 +/- 60
27th
3
1742 +/- 21
27th
3
1741 +/- 6
26th
4
1716 +/- 12
Low-Tier
25th
< 1712
25th
4
1699 +/- 3
24th
9
1622 +/- 42
24th
5
1602 +/- 7
23rd
4
1549 +/- 9
13th
6
1429 +/- 2
13th
3
1415 +/- 169
12th
3
1408 +/- 5
12th
6
1394 +/- 3
11th
6
1386 +/- 4
11th
4
1381 +/- 18
10th
4
1353 +/- 17
10th
3
1324 +/- 11
9th
19
1282 +/- 34
7th
14
1232 +/- 74
6th
4
1202 +/- 12
6th
3
1164 +/- 8
5th
6
1132 +/- 4
4th
12
1006 +/- 45
3rd
4
905 +/- 21
3rd
3
884 +/- 1
2nd
3
554 +/- 60
OpenBenchmarking.orgDistribution Of Public Results - Device AI Score1195 Results Range From 243 To 22485 Score24368811331578202324682913335838034248469351385583602864736918736378088253869891439588100331047810923113681181312258127031314813593140381448314928153731581816263167081715317598180431848818933193781982320268207132115821603220482249360120180240300

Based on OpenBenchmarking.org data, the selected test / test configuration (AI Benchmark Alpha 0.1.2 - Device AI Score) has an average run-time of 21 minutes. By default this test profile is set to run at least 1 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 AI ScoreRun-Time714212835Min: 16 / Avg: 20.95 / Max: 32

Does It Scale Well With Increasing Cores?

Yes, based on the automated analysis of the collected public benchmark data, this test / test settings does 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 BaseAI Benchmark Alpha CPU Core ScalingDevice AI Score4681216243248641281.09792.19583.29374.39165.4895