fgds

Intel Core i5-12600K testing with a ASUS PRIME Z690-P WIFI D4 (0605 BIOS) and ASUS Intel ADL-S GT1 15GB on Ubuntu 22.04 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2210141-PTS-FGDS239097
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
A
October 14 2022
  1 Hour, 37 Minutes
B
October 14 2022
  1 Hour, 36 Minutes
C
October 14 2022
  1 Hour, 35 Minutes
Invert Behavior (Only Show Selected Data)
  1 Hour, 36 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


fgdsOpenBenchmarking.orgPhoronix Test SuiteIntel Core i5-12600K @ 6.30GHz (10 Cores / 16 Threads)ASUS PRIME Z690-P WIFI D4 (0605 BIOS)Intel Device 7aa716GB1000GB Western Digital WDS100T1X0E-00AFY0ASUS Intel ADL-S GT1 15GB (1450MHz)Realtek ALC897ASUS MG28URealtek RTL8125 2.5GbE + Intel Device 7af0Ubuntu 22.045.19.0-051900rc6daily20220716-generic (x86_64)GNOME Shell 42.1X Server 1.21.1.3 + Wayland4.6 Mesa 22.0.1OpenCL 3.01.2.204GCC 11.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLOpenCLVulkanCompilerFile-SystemScreen ResolutionFgds PerformanceSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9 - Python 3.10.4- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

ABCResult OverviewPhoronix Test Suite100%101%102%103%OpenRadiossAOM AV1oneDNNspaCyQuadRayTensorFlowY-CruncherNeural Magic DeepSparseSMHasher

fgdsonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUaom-av1: Speed 10 Realtime - Bosphorus 1080ponednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUaom-av1: Speed 6 Realtime - Bosphorus 1080popenradioss: Bumper Beamonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUopenradioss: Cell Phone Drop Testaom-av1: Speed 8 Realtime - Bosphorus 1080pquadray: 5 - 1080ponednn: IP Shapes 3D - u8s8f32 - CPUaom-av1: Speed 4 Two-Pass - Bosphorus 1080paom-av1: Speed 9 Realtime - Bosphorus 1080paom-av1: Speed 6 Two-Pass - Bosphorus 1080popenradioss: Bird Strike on Windshieldaom-av1: Speed 8 Realtime - Bosphorus 4Kdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamonednn: IP Shapes 1D - f32 - CPUopenradioss: INIVOL and Fluid Structure Interaction Drop Containeronednn: Recurrent Neural Network Training - f32 - CPUaom-av1: Speed 6 Realtime - Bosphorus 4Konednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUaom-av1: Speed 9 Realtime - Bosphorus 4Kdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamy-cruncher: 500Mdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamspacy: en_core_web_trfaom-av1: Speed 10 Realtime - Bosphorus 4Kdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamy-cruncher: 1Bonednn: Recurrent Neural Network Training - u8s8f32 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamaom-av1: Speed 0 Two-Pass - Bosphorus 1080pdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamopenradioss: Rubber O-Ring Seal Installationdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamsmhasher: MeowHash x86_64 AES-NIdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamonednn: Recurrent Neural Network Inference - u8s8f32 - CPUtensorflow: CPU - 32 - AlexNetdeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamaom-av1: Speed 6 Two-Pass - Bosphorus 4Konednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUaom-av1: Speed 4 Two-Pass - Bosphorus 4Konednn: Deconvolution Batch shapes_3d - f32 - CPUdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUspacy: en_core_web_lgonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUsmhasher: FarmHash32 x86_64 AVXtensorflow: CPU - 32 - GoogLeNetdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamtensorflow: CPU - 16 - ResNet-50tensorflow: CPU - 16 - GoogLeNettensorflow: CPU - 32 - ResNet-50smhasher: SHA3-256quadray: 1 - 1080pdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamsmhasher: wyhashquadray: 3 - 1080pquadray: 1 - 4Kquadray: 2 - 1080psmhasher: FarmHash128onednn: IP Shapes 1D - u8s8f32 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamtensorflow: CPU - 16 - AlexNetsmhasher: Spooky32smhasher: t1ha2_atoncesmhasher: t1ha0_aes_avx2 x86_64smhasher: fasthash32aom-av1: Speed 0 Two-Pass - Bosphorus 4Kquadray: 5 - 4Kquadray: 3 - 4Kquadray: 2 - 4Ksmhasher: MeowHash x86_64 AES-NIsmhasher: t1ha0_aes_avx2 x86_64smhasher: FarmHash32 x86_64 AVXsmhasher: t1ha2_atoncesmhasher: FarmHash128smhasher: fasthash32smhasher: Spooky32smhasher: SHA3-256smhasher: wyhashABC1.15139152.852.8157358.85247.093.21457176.92120.561.252.1633216.65143.7440.35385.8247.8141.02624.37284.02417801.994508.3232.469.8148512.323366.856.68636.8952181.391587.706520.709174.6769108967.827.449220.491548.792643.3964474.24154.38736.47710.7128.6104288.06152.31496.565249353.69718.17842168.08102.2838.352726.067744.109712.812161.757.148.14523112.51482.0929710.60492177.71722614.32554439.6426888.8354.4356.779287.761317.8352.5218.44244.0719.56.893570.860714.10822.928643.606329154.234.594.975.619170.141.46342719.456480.0417969.0121936.5572005.917604.240.240.321.181.4254.28325.06532.83425.29653.74228.18633.8531599.28118.5521.01232178.372.7458957.88229.962.97401167.33127.41.32.283217.49149.2541.89372.3449.4441.387124.16023.91825783.864410.3133.169.8199712.445567.4756.79346.9928180.398187.681520.576175.7057107068.6727.585320.571248.603943.4694409.87154.53266.4710.7228.3942285.73152.89516.540348796.81713.09472159.81102.9538.266926.124944.233612.872160.417.178.06747112.25592.1116710.69942160.581725414.43414407.5826897.9854.0656.41688.319717.8852.2518.39245.0119.576.879370.740314.131922.913643.634829184.544.64.985.6119201.691.46173720.353380.117981.7121929.7772004.317603.850.240.321.181.4254.3325.06532.60825.29653.74228.18633.7821595.17618.5521.30001153.462.5492163.46228.073.14484164.08125.131.322.2760217.49149.7441.88375.1249.0340.162324.89683.98114785.094439.8932.910.026512.185168.1555.6856.8642183.736789.275420.343172.6192107468.9627.162120.807248.053344.0534439.06152.3166.56510.7228.763284.58151.13116.616649349.97720.94022183.19103.3437.96126.336343.783612.942182.097.218.11715113.28162.0937710.69772179.571711814.39224429.5426713.6854.2956.740688.067417.9352.3718.47245.0819.496.870670.659214.148122.868243.721129220.874.64.975.6119197.591.46136720.177480.0417979.5621936.7371999.167604.380.240.321.181.4254.31525.06532.82325.29653.74228.18833.8181596.08318.551OpenBenchmarking.org

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUABC0.29250.5850.87751.171.46251.151391.012321.30001MIN: 1.06MIN: 0.93MIN: 1.151. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pABC4080120160200152.85178.37153.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUABC0.63351.2671.90052.5343.16752.815732.745892.54921MIN: 2.72MIN: 2.68MIN: 2.481. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pABC142842567058.8557.8863.461. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper BeamABC50100150200250247.09229.96228.07

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUABC0.72331.44662.16992.89323.61653.214572.974013.14484MIN: 3.16MIN: 2.95MIN: 3.091. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop TestABC4080120160200176.92167.33164.08

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pABC306090120150120.56127.40125.131. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 1080pABC0.2970.5940.8911.1881.4851.251.301.321. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUABC0.51371.02741.54112.05482.56852.163322.283202.27602MIN: 2.13MIN: 2.25MIN: 2.241. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pABC4812162016.6517.4917.491. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pABC306090120150143.74149.25149.741. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pABC102030405040.3541.8941.881. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on WindshieldABC80160240320400385.82372.34375.12

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KABC112233445547.8149.4449.031. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamABC91827364541.0341.3940.16

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamABC61218243024.3724.1624.90

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUABC0.90541.81082.71623.62164.5274.024173.918253.98114MIN: 3.88MIN: 3.77MIN: 3.821. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop ContainerABC2004006008001000801.99783.86785.09

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUABC100020003000400050004508.324410.314439.89MIN: 4494.66MIN: 4402.29MIN: 4432.351. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KABC81624324032.4633.1632.901. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUABC36912159.814859.8199710.02650MIN: 6.1MIN: 6.62MIN: 6.351. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUABC369121512.3212.4512.19MIN: 12.01MIN: 12.3MIN: 12.061. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KABC153045607566.8067.4768.151. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamABC132639526556.6956.7955.69

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamABC2468106.89526.99286.8642

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamABC4080120160200181.39180.40183.74

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamABC2040608010087.7187.6889.28

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 500MABC51015202520.7120.5820.34

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamABC4080120160200174.68175.71172.62

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_trfABC2004006008001000108910701074

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4KABC153045607567.8068.6768.961. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamABC61218243027.4527.5927.16

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamABC51015202520.4920.5720.81

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamABC112233445548.7948.6048.05

Y-Cruncher

Y-Cruncher is a multi-threaded Pi benchmark capable of computing Pi to trillions of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterY-Cruncher 0.7.10.9513Pi Digits To Calculate: 1BABC102030405043.4043.4744.05

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUABC100020003000400050004474.244409.874439.06MIN: 4465.82MIN: 4403.1MIN: 4431.621. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamABC306090120150154.39154.53152.32

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamABC2468106.47716.47106.5651

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pABC0.1620.3240.4860.6480.810.710.720.721. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamABC71421283528.6128.3928.76

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal InstallationABC60120180240300288.06285.73284.58

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamABC306090120150152.31152.90151.13

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamABC2468106.56526.54036.6166

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIABC11K22K33K44K55K49353.6948796.8149349.971. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamABC160320480640800718.18713.09720.94

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUABC50010001500200025002168.082159.812183.19MIN: 2161.12MIN: 2154.11MIN: 2177.611. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNetABC20406080100102.28102.95103.34

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-StreamABC91827364538.3538.2737.96

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-StreamABC61218243026.0726.1226.34

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-StreamABC102030405044.1144.2343.78

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4KABC369121512.8112.8712.941. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUABC50010001500200025002161.752160.412182.09MIN: 2154.79MIN: 2153.51MIN: 2174.631. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4KABC2468107.147.177.211. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUABC2468108.145238.067478.11715MIN: 8.1MIN: 8.02MIN: 8.041. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-StreamABC306090120150112.51112.26113.28

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUABC0.47510.95021.42531.90042.37552.092972.111672.09377MIN: 2.06MIN: 2.03MIN: 2.051. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUABC369121510.6010.7010.70MIN: 10.49MIN: 10.59MIN: 10.581. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUABC50010001500200025002177.702160.582179.57MIN: 2166.44MIN: 2154.5MIN: 21731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgtokens/sec, More Is BetterspaCy 3.4.1Model: en_core_web_lgABC4K8K12K16K20K172261725417118

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUABC4812162014.3314.4314.39MIN: 14.19MIN: 14.23MIN: 14.21. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUABC100020003000400050004439.644407.584429.54MIN: 4434.42MIN: 4400MIN: 4422.181. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXABC6K12K18K24K30K26888.8326897.9826713.681. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNetABC122436486054.4354.0654.29

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamABC132639526556.7856.4256.74

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamABC2040608010087.7688.3288.07

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: ResNet-50ABC4812162017.8317.8817.93

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNetABC122436486052.5252.2552.37

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: ResNet-50ABC51015202518.4418.3918.47

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: SHA3-256ABC50100150200250244.07245.01245.081. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 1080pABC51015202519.5019.5719.491. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreamABC2468106.89356.87936.8706

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreamABC163248648070.8670.7470.66

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreamABC4812162014.1114.1314.15

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreamABC51015202522.9322.9122.87

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreamABC102030405043.6143.6343.72

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: wyhashABC6K12K18K24K30K29154.2329184.5429220.871. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 1080pABC1.0352.073.1054.145.1754.594.604.601. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 1 - Resolution: 4KABC1.12052.2413.36154.4825.60254.974.984.971. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 1080pABC1.26232.52463.78695.04926.31155.605.615.611. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash128ABC4K8K12K16K20K19170.1419201.6919197.591. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.7Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUABC0.32930.65860.98791.31721.64651.463421.461731.46136MIN: 1.44MIN: 1.44MIN: 1.441. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreamABC160320480640800719.46720.35720.18

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNetABC2040608010080.0480.1080.04

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: Spooky32ABC4K8K12K16K20K17969.0117981.7117979.561. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha2_atonceABC5K10K15K20K25K21936.5521929.7721936.731. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64ABC15K30K45K60K75K72005.9172004.3171999.161. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: fasthash32ABC160032004800640080007604.247603.857604.381. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4KABC0.0540.1080.1620.2160.270.240.240.241. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

QuadRay

VectorChief's QuadRay is a real-time ray-tracing engine written to support SIMD across ARM, MIPS, PPC, and x86/x86_64 processors. QuadRay supports SSE/SSE2/SSE4 and AVX/AVX2/AVX-512 usage on Intel/AMD CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 5 - Resolution: 4KABC0.0720.1440.2160.2880.360.320.320.321. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 3 - Resolution: 4KABC0.26550.5310.79651.0621.32751.181.181.181. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

OpenBenchmarking.orgFPS, More Is BetterQuadRay 2022.05.25Scene: 2 - Resolution: 4KABC0.31950.6390.95851.2781.59751.421.421.421. (CXX) g++ options: -O3 -pthread -lm -lstdc++ -lX11 -lXext -lpthread

92 Results Shown

oneDNN
AOM AV1
oneDNN
AOM AV1
OpenRadioss
oneDNN
OpenRadioss
AOM AV1
QuadRay
oneDNN
AOM AV1:
  Speed 4 Two-Pass - Bosphorus 1080p
  Speed 9 Realtime - Bosphorus 1080p
  Speed 6 Two-Pass - Bosphorus 1080p
OpenRadioss
AOM AV1
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream:
    ms/batch
    items/sec
oneDNN
OpenRadioss
oneDNN
AOM AV1
oneDNN:
  Deconvolution Batch shapes_1d - f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
AOM AV1
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
Y-Cruncher
Neural Magic DeepSparse
spaCy
AOM AV1
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream
Y-Cruncher
oneDNN
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
AOM AV1
Neural Magic DeepSparse
OpenRadioss
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    ms/batch
    items/sec
SMHasher
Neural Magic DeepSparse
oneDNN
TensorFlow
Neural Magic DeepSparse:
  CV Detection,YOLOv5s COCO - Synchronous Single-Stream:
    items/sec
    ms/batch
  CV Detection,YOLOv5s COCO - Asynchronous Multi-Stream:
    items/sec
AOM AV1
oneDNN
AOM AV1
oneDNN
Neural Magic DeepSparse
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  IP Shapes 3D - f32 - CPU
  Recurrent Neural Network Inference - f32 - CPU
spaCy
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
SMHasher
TensorFlow
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    ms/batch
    items/sec
TensorFlow:
  CPU - 16 - ResNet-50
  CPU - 16 - GoogLeNet
  CPU - 32 - ResNet-50
SMHasher
QuadRay
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream
SMHasher
QuadRay:
  3 - 1080p
  1 - 4K
  2 - 1080p
SMHasher
oneDNN
Neural Magic DeepSparse
TensorFlow
SMHasher:
  Spooky32
  t1ha2_atonce
  t1ha0_aes_avx2 x86_64
  fasthash32
AOM AV1
QuadRay:
  5 - 4K
  3 - 4K
  2 - 4K