n1n1

ARMv8 Neoverse-N1 testing with a GIGABYTE G242-P36-00 MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS) and ASPEED on Ubuntu 23.10 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 2403174-NE-N1N13670960
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

Limit displaying results to tests within:

C/C++ Compiler Tests 2 Tests
CPU Massive 6 Tests
Creator Workloads 7 Tests
Encoding 2 Tests
HPC - High Performance Computing 3 Tests
Imaging 2 Tests
Machine Learning 3 Tests
Multi-Core 7 Tests
Intel oneAPI 2 Tests
Python Tests 2 Tests
Server CPU Tests 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
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
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a
March 17
  15 Minutes
aa
March 17
  7 Hours, 43 Minutes
b
March 17
  2 Hours, 32 Minutes
c
March 17
  2 Hours, 15 Minutes
Invert Hiding All Results Option
  3 Hours, 11 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):

a

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

aa

b

c

Processor: ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores), Motherboard: GIGABYTE G242-P36-00 MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS), Chipset: Ampere Computing LLC Altra PCI Root Complex A, Memory: 16 x 32 GB DDR4-3200MT/s Samsung M393A4K40DB3-CWE, Disk: 800GB Micron_7450_MTFDKBA800TFS, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Intel I350

OS: Ubuntu 23.10, Kernel: 6.5.0-15-generic (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1024x768

n1n1 OpenBenchmarking.orgPhoronix Test SuiteARMv8 Neoverse-N1 @ 3.00GHz (128 Cores)GIGABYTE G242-P36-00 MP32-AR2-00 v01000100 (F31k SCPAmpere Computing LLC Altra PCI Root Complex A16 x 32 GB DDR4-3200MT/s Samsung M393A4K40DB3-CWE800GB Micron_7450_MTFDKBA800TFSASPEEDVGA HDMI2 x Intel I350Ubuntu 23.106.5.0-15-generic (aarch64)GCC 13.2.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelCompilerFile-SystemScreen ResolutionN1n1 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - Scaling Governor: cppc_cpufreq performance (Boost: Disabled)- Python 3.11.6- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

aaabcResult OverviewPhoronix Test Suite100%104%107%111%115%StockfishJPEG-XL Decoding libjxlJPEG-XL libjxlTimed Linux Kernel CompilationSVT-AV1

n1n1 jpegxl-decode: Alljpegxl: PNG - 80jpegxl: JPEG - 90jpegxl: PNG - 90jpegxl: JPEG - 80draco: Church Facadeonednn: Deconvolution Batch shapes_1d - CPUdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streambuild-linux-kernel: defconfigopenvino: Face Detection Retail FP16 - CPUjpegxl: JPEG - 100openvino: Face Detection Retail FP16 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUprimesieve: 1e12jpegxl: PNG - 100openvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamsrsran: PDSCH Processor Benchmark, Throughput Totaldeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamcompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img Compressiondeepsparse: Llama2 Chat 7b Quantized - Asynchronous Multi-Streamonednn: IP Shapes 1D - CPUjpegxl-decode: 1openvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUsvt-av1: Preset 12 - Bosphorus 4Kdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamopenvino: Person Detection FP32 - CPUdeepsparse: Llama2 Chat 7b Quantized - Asynchronous Multi-Streamonednn: Deconvolution Batch shapes_3d - CPUopenvino: Person Detection FP32 - CPUdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streambuild-linux-kernel: allmodconfigsvt-av1: Preset 8 - Bosphorus 1080ponednn: Recurrent Neural Network Inference - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamopenvino: Handwritten English Recognition FP16-INT8 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamsvt-av1: Preset 12 - Bosphorus 1080pdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamdeepsparse: Llama2 Chat 7b Quantized - Synchronous Single-Streamopenvino: Vehicle Detection FP16 - CPUsvt-av1: Preset 13 - Bosphorus 1080popenvino: Vehicle Detection FP16 - CPUdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamsvt-av1: Preset 13 - Bosphorus 4Kdraco: Liondeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamopenvino: Face Detection FP16-INT8 - CPUprimesieve: 1e13onednn: Convolution Batch Shapes Auto - CPUopenvino: Person Detection FP16 - CPUonednn: IP Shapes 3D - CPUopenvino: Face Detection FP16-INT8 - CPUsvt-av1: Preset 8 - Bosphorus 4Kopenvino: Road Segmentation ADAS FP16-INT8 - CPUdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamopenvino: Vehicle Detection FP16-INT8 - CPUsvt-av1: Preset 4 - Bosphorus 4Kdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamopenvino: Person Detection FP16 - CPUdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamopenvino: Road Segmentation ADAS FP16-INT8 - CPUdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Face Detection FP16 - CPUsvt-av1: Preset 4 - Bosphorus 1080popenvino: Handwritten English Recognition FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUsrsran: PDSCH Processor Benchmark, Throughput Threadopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Noise Suppression Poconet-Like FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUopenvino: Person Re-Identification Retail FP16 - CPUonednn: Recurrent Neural Network Training - CPUencode-wavpack: WAV To WavPackopenvino: Face Detection FP16 - CPUsrsran: PUSCH Processor Benchmark, Throughput Threadsrsran: PUSCH Processor Benchmark, Throughput Totalstockfish: Chess Benchmarkaaabc558.56943.09737.59139.24939.26894.27331.66529.60314099.827.23774.68256.897265.743364.39974.89624.9452.6528.914175.846.71602.159028775523.01940.27937.41537.89538.9211010020.92552678.238223.5039182.8789345.108092.760676.5931.12147.28217.95146.719.0819109.952326.087138.3162108.97293.472.91129.238204.69156.22315.74503.150813936.17.5520132.1849132.9525474.89762.4135532.26024.8406527.152333.1595.9874.46933.41871462.9421.86202.63592156.8721332.89312.7962614.73310.5129348.01857.1351460.94147.761844.1246216.1826.253138.0726264.978132.542512.929877.3026143.42363.354222.86476.355774.900735155.02611149.472430.596132.659733.5337133.53017.47412.7442.3054.294702150.302.1558211232.4324.92734.90438.7131357.862.6441337.5918112.53348.87091840.367714.7750.597619.7459794.0640.11143.831746.6120913.417.4691133.621889.3565.60486.1110877.538.925194.88163.9522.801402.51175.7193.84164.82142.60224.223738.3925.1992.8459449725564.89341.30937.7939.66937.766984720.43082688.956723.4116181.8956346.669994.426664.7831.62148.1221.47144.389.1198109.478425.945338.5246107.5297.482.87229.494205.33155.74312.46333.183513999.67.5061132.9867132.7356475.82122.4393382.27544.8801527.417329.9796.975.16733.71251473.2321.71202.14712140.221231.56412.7823814.84311.1676350.29457.0271461147.081834.8257217.1626.346837.9374265.435131.952912.897777.4941142.79365.102223.85478.641874.958732055.25551144.801230.721132.527233.5843134.00167.44842.7542.4414.280362151.852.1517811206.1325.00634.79439.6023358.52.651335.3456112.82918.84831835.257214.7750.732819.6933793.3140.15143.483746.715915.177.4692133.625389.1965.49486.910891.938.921194.84163.9822.791402.97193.93164.75142.58224.223737.1525.2052.8451901853542.10341.35435.84339.25139.315984820.89252630.33423.9126185.7196339.994.496670.1931.62447.72219.27145.828.982111.157826.331537.9597108.56294.582.89329.544207.24154.3316.3473.14497.5924131.4797131.5237479.99012.4386312.28364.8885827.396331.7796.3875.01533.67971460.7221.89201.02442146.0721169.29532.8038614.8312.7909349.91556.7891469.65146.91833.4487217.4126.200238.1495264.28131.859412.960577.119143.48363.612222.78478.373274.604733255.24351144.772730.667532.583533.6663133.48667.47672.7542.2944.284612157.452.1487811196.5424.95234.88438.2501357.412.651333.7177112.85218.84611836.79314.7350.64919.726279240.21143.602546.6799913.217.454133.885389.365.6486.1110876.78.926194.65164.1322.781403.65193.87164.79142.54224.313738.5325.22.8453514996OpenBenchmarking.org

JPEG-XL Decoding libjxl

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: Allcbaaa120240360480600SE +/- 1.96, N = 3542.10564.89523.02558.57
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: Allcbaaa100200300400500Min: 519.16 / Avg: 523.02 / Max: 525.52

JPEG-XL libjxl

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 80cbaaa1020304050SE +/- 0.30, N = 341.3541.3140.2843.101. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 80cbaaa918273645Min: 39.71 / Avg: 40.28 / Max: 40.741. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 90cbaaa918273645SE +/- 0.45, N = 1535.8437.7937.4237.591. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 90cbaaa816243240Min: 35.16 / Avg: 37.42 / Max: 40.381. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 90cbaaa918273645SE +/- 0.55, N = 1539.2539.6737.9039.251. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 90cbaaa816243240Min: 32.82 / Avg: 37.9 / Max: 39.131. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 80cbaaa918273645SE +/- 0.12, N = 339.3237.7738.9239.271. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 80cbaaa816243240Min: 38.75 / Avg: 38.92 / Max: 39.151. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

Google Draco

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church Facadecbaa2K4K6K8K10KSE +/- 6.24, N = 398489847101001. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church Facadecbaa2K4K6K8K10KMin: 10088 / Avg: 10100 / Max: 101091. (CXX) g++ options: -O3

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_1d - Engine: CPUcbaa510152025SE +/- 0.20, N = 320.8920.4320.93MIN: 19.81MIN: 19.32MIN: 19.341. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_1d - Engine: CPUcbaa510152025Min: 20.54 / Avg: 20.93 / Max: 21.181. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa6001200180024003000SE +/- 6.53, N = 32630.332688.962678.24
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa5001000150020002500Min: 2666.07 / Avg: 2678.24 / Max: 2688.41

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa612182430SE +/- 0.06, N = 323.9123.4123.50
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa612182430Min: 23.4 / Avg: 23.5 / Max: 23.59

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamcbaa4080120160200SE +/- 0.08, N = 3185.72181.90182.88
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamcbaa306090120150Min: 182.74 / Avg: 182.88 / Max: 183.03

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamcbaa80160240320400SE +/- 0.25, N = 3339.90346.67345.11
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamcbaa60120180240300Min: 344.66 / Avg: 345.11 / Max: 345.52

Timed Linux Kernel Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfigcbaaa20406080100SE +/- 0.90, N = 394.5094.4392.7694.27
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfigcbaaa20406080100Min: 91.85 / Avg: 92.76 / Max: 94.57

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUcbaa150300450600750SE +/- 8.52, N = 3670.19664.78676.591. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUcbaa120240360480600Min: 660.75 / Avg: 676.59 / Max: 689.941. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

JPEG-XL libjxl

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 100cbaaa714212835SE +/- 0.00, N = 331.6231.6231.1231.671. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: JPEG - Quality: 100cbaaa714212835Min: 31.12 / Avg: 31.12 / Max: 31.131. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUcbaa1122334455SE +/- 0.60, N = 347.7248.1047.28MIN: 9.97 / MAX: 99.86MIN: 9.92 / MAX: 115.12MIN: 10.17 / MAX: 121.041. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16 - Device: CPUcbaa1020304050Min: 46.35 / Avg: 47.28 / Max: 48.41. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUcbaa50100150200250SE +/- 0.18, N = 3219.27221.47217.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUcbaa4080120160200Min: 217.6 / Avg: 217.95 / Max: 218.191. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUcbaa306090120150SE +/- 0.12, N = 3145.82144.38146.71MIN: 96.38 / MAX: 1563.28MIN: 96.65 / MAX: 1566.66MIN: 96.02 / MAX: 1572.431. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16-INT8 - Device: CPUcbaa306090120150Min: 146.56 / Avg: 146.71 / Max: 146.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamcbaa3691215SE +/- 0.0713, N = 38.98209.11989.0819
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamcbaa3691215Min: 8.99 / Avg: 9.08 / Max: 9.22

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamcbaa20406080100SE +/- 0.86, N = 3111.16109.48109.95
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamcbaa20406080100Min: 108.27 / Avg: 109.95 / Max: 111.08

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamcbaa612182430SE +/- 0.11, N = 326.3325.9526.09
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamcbaa612182430Min: 25.92 / Avg: 26.09 / Max: 26.29

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamcbaa918273645SE +/- 0.16, N = 337.9638.5238.32
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamcbaa816243240Min: 38.01 / Avg: 38.32 / Max: 38.56

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPUcbaa20406080100SE +/- 0.11, N = 3108.56107.50108.97MIN: 17.21 / MAX: 1188.34MIN: 57.15 / MAX: 1202.08MIN: 17.48 / MAX: 1207.621. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPUcbaa20406080100Min: 108.84 / Avg: 108.97 / Max: 109.191. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPUcbaa60120180240300SE +/- 0.30, N = 3294.58297.48293.471. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Weld Porosity Detection FP16 - Device: CPUcbaa50100150200250Min: 292.87 / Avg: 293.47 / Max: 293.811. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Primesieve

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12cbaa0.6551.311.9652.623.275SE +/- 0.003, N = 32.8932.8722.9111. (CXX) g++ options: -O3
OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e12cbaa246810Min: 2.91 / Avg: 2.91 / Max: 2.921. (CXX) g++ options: -O3

JPEG-XL libjxl

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 100cbaaa714212835SE +/- 0.04, N = 329.5429.4929.2429.601. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL libjxl 0.10.1Input: PNG - Quality: 100cbaaa714212835Min: 29.15 / Avg: 29.24 / Max: 29.31. (CXX) g++ options: -fno-rtti -O3 -fPIE -pie -lm

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUcbaa50100150200250SE +/- 0.66, N = 3207.24205.33204.691. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUcbaa4080120160200Min: 203.9 / Avg: 204.69 / Max: 205.991. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUcbaa306090120150SE +/- 0.49, N = 3154.30155.74156.22MIN: 44.57 / MAX: 239.56MIN: 48.23 / MAX: 240.13MIN: 44.3 / MAX: 240.551. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Vehicle Bike Detection FP16 - Device: CPUcbaa306090120150Min: 155.24 / Avg: 156.22 / Max: 156.821. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa70140210280350SE +/- 0.75, N = 3316.35312.46315.75
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa60120180240300Min: 314.29 / Avg: 315.74 / Max: 316.81

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa0.71631.43262.14892.86523.5815SE +/- 0.0074, N = 33.14493.18353.1508
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa246810Min: 3.14 / Avg: 3.15 / Max: 3.17

srsRAN Project

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PDSCH Processor Benchmark, Throughput Totalbaaa3K6K9K12K15KSE +/- 42.60, N = 313999.613936.114099.81. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl
OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PDSCH Processor Benchmark, Throughput Totalbaaa2K4K6K8K10KMin: 13854.9 / Avg: 13936.07 / Max: 13999.11. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa246810SE +/- 0.0154, N = 37.59247.50617.5520
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa3691215Min: 7.52 / Avg: 7.55 / Max: 7.57

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa306090120150SE +/- 0.27, N = 3131.48132.99132.18
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa20406080100Min: 131.83 / Avg: 132.18 / Max: 132.71

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamcbaa306090120150SE +/- 0.39, N = 3131.52132.74132.95
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamcbaa20406080100Min: 132.43 / Avg: 132.95 / Max: 133.72

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamcbaa100200300400500SE +/- 1.33, N = 3479.99475.82474.90
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamcbaa90180270360450Min: 472.43 / Avg: 474.9 / Max: 476.97

Parallel BZIP2 Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img Compressioncbaa0.54891.09781.64672.19562.7445SE +/- 0.001512, N = 32.4386312.4393382.4135531. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img Compressioncbaa246810Min: 2.41 / Avg: 2.41 / Max: 2.421. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamcbaa0.51381.02761.54142.05522.569SE +/- 0.0074, N = 32.28362.27542.2602
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamcbaa246810Min: 2.25 / Avg: 2.26 / Max: 2.27

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 1D - Engine: CPUcbaa1.09992.19983.29974.39965.4995SE +/- 0.01022, N = 34.888584.880154.84065MIN: 4.3MIN: 4.23MIN: 4.251. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 1D - Engine: CPUcbaa246810Min: 4.83 / Avg: 4.84 / Max: 4.861. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

JPEG-XL Decoding libjxl

OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: 1cbaaa612182430SE +/- 0.01, N = 327.4027.4227.1527.24
OpenBenchmarking.orgMP/s, More Is BetterJPEG-XL Decoding libjxl 0.10.1CPU Threads: 1cbaaa612182430Min: 27.13 / Avg: 27.15 / Max: 27.18

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUcbaa70140210280350SE +/- 0.61, N = 3331.77329.97333.151. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUcbaa60120180240300Min: 331.94 / Avg: 333.15 / Max: 333.941. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUcbaa20406080100SE +/- 0.18, N = 396.3896.9095.98MIN: 69.36 / MAX: 140.93MIN: 70.14 / MAX: 141.32MIN: 71.43 / MAX: 140.321. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection Retail FP16-INT8 - Device: CPUcbaa20406080100Min: 95.74 / Avg: 95.98 / Max: 96.331. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 4Kcbaaa20406080100SE +/- 0.28, N = 375.0275.1774.4774.681. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 4Kcbaaa1428425670Min: 73.92 / Avg: 74.47 / Max: 74.891. (CXX) g++ options: -march=native

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamcbaa816243240SE +/- 0.02, N = 333.6833.7133.42
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamcbaa714212835Min: 33.37 / Avg: 33.42 / Max: 33.45

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUcbaa30060090012001500SE +/- 1.48, N = 31460.721473.231462.941. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUcbaa30060090012001500Min: 1460.11 / Avg: 1462.94 / Max: 1465.121. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUcbaa510152025SE +/- 0.02, N = 321.8921.7121.86MIN: 2.07 / MAX: 156.71MIN: 2.05 / MAX: 156.88MIN: 2 / MAX: 157.11. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUcbaa510152025Min: 21.83 / Avg: 21.86 / Max: 21.91. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa4080120160200SE +/- 0.34, N = 3201.02202.15202.64
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa4080120160200Min: 201.98 / Avg: 202.64 / Max: 203.07

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPUcbaa5001000150020002500SE +/- 2.39, N = 32146.072140.202156.87MIN: 439.17 / MAX: 2969.83MIN: 527.18 / MAX: 2951.37MIN: 504.09 / MAX: 29901. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPUcbaa400800120016002000Min: 2152.53 / Avg: 2156.87 / Max: 2160.771. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamcbaa5K10K15K20K25KSE +/- 55.68, N = 321169.3021231.5621332.89
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Streamcbaa4K8K12K16K20KMin: 21228.1 / Avg: 21332.89 / Max: 21417.91

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_3d - Engine: CPUcbaa0.63091.26181.89272.52363.1545SE +/- 0.01912, N = 122.803862.782382.79626MIN: 2.7MIN: 2.72MIN: 2.681. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Deconvolution Batch shapes_3d - Engine: CPUcbaa246810Min: 2.75 / Avg: 2.8 / Max: 31. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPUcbaa48121620SE +/- 0.02, N = 314.8014.8414.731. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP32 - Device: CPUcbaa48121620Min: 14.7 / Avg: 14.73 / Max: 14.761. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa70140210280350SE +/- 0.51, N = 3312.79311.17310.51
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa60120180240300Min: 309.92 / Avg: 310.51 / Max: 311.53

Timed Linux Kernel Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfigcbaa80160240320400SE +/- 0.68, N = 3349.92350.29348.02
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfigcbaa60120180240300Min: 347.14 / Avg: 348.02 / Max: 349.36

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 1080pcbaaa1326395265SE +/- 0.06, N = 356.7957.0357.1456.901. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 1080pcbaaa1122334455Min: 57.02 / Avg: 57.14 / Max: 57.221. (CXX) g++ options: -march=native

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Inference - Engine: CPUcbaa30060090012001500SE +/- 3.72, N = 31469.651461.001460.94MIN: 1448.43MIN: 1442.49MIN: 1436.361. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Inference - Engine: CPUcbaa30060090012001500Min: 1455.38 / Avg: 1460.94 / Max: 14681. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUcbaa306090120150SE +/- 0.83, N = 3146.90147.08147.761. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUcbaa306090120150Min: 146.9 / Avg: 147.76 / Max: 149.421. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamcbaa400800120016002000SE +/- 1.78, N = 31833.451834.831844.12
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamcbaa30060090012001500Min: 1840.8 / Avg: 1844.12 / Max: 1846.89

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUcbaa50100150200250SE +/- 1.21, N = 3217.41217.16216.18MIN: 210.44 / MAX: 372.96MIN: 208.82 / MAX: 374.93MIN: 206.9 / MAX: 376.91. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16-INT8 - Device: CPUcbaa4080120160200Min: 213.75 / Avg: 216.18 / Max: 217.451. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamcbaa612182430SE +/- 0.02, N = 326.2026.3526.25
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamcbaa612182430Min: 26.21 / Avg: 26.25 / Max: 26.3

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamcbaa918273645SE +/- 0.04, N = 338.1537.9438.07
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamcbaa816243240Min: 38.01 / Avg: 38.07 / Max: 38.14

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 1080pcbaaa60120180240300SE +/- 0.05, N = 3264.28265.44264.98265.741. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 12 - Input: Bosphorus 1080pcbaaa50100150200250Min: 264.89 / Avg: 264.98 / Max: 265.061. (CXX) g++ options: -march=native

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamcbaa306090120150SE +/- 0.34, N = 3131.86131.95132.54
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamcbaa20406080100Min: 131.9 / Avg: 132.54 / Max: 133.03

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamcbaa3691215SE +/- 0.02, N = 312.9612.9012.93
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamcbaa48121620Min: 12.91 / Avg: 12.93 / Max: 12.97

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamcbaa20406080100SE +/- 0.12, N = 377.1277.4977.30
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Streamcbaa1530456075Min: 77.06 / Avg: 77.3 / Max: 77.45

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPUcbaa306090120150SE +/- 0.06, N = 3143.48142.79143.42MIN: 44.55 / MAX: 252.93MIN: 60 / MAX: 245.21MIN: 62.82 / MAX: 295.21. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPUcbaa306090120150Min: 143.35 / Avg: 143.42 / Max: 143.541. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 1080pcbaaa80160240320400SE +/- 0.57, N = 3363.61365.10363.35364.401. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 1080pcbaaa70140210280350Min: 362.21 / Avg: 363.35 / Max: 363.991. (CXX) g++ options: -march=native

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPUcbaa50100150200250SE +/- 0.10, N = 3222.78223.85222.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16 - Device: CPUcbaa4080120160200Min: 222.67 / Avg: 222.86 / Max: 222.981. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamcbaa100200300400500SE +/- 1.18, N = 3478.37478.64476.36
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamcbaa80160240320400Min: 474.77 / Avg: 476.36 / Max: 478.67

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4Kcbaaa20406080100SE +/- 0.19, N = 374.6074.9674.9074.901. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 13 - Input: Bosphorus 4Kcbaaa1428425670Min: 74.57 / Avg: 74.9 / Max: 75.221. (CXX) g++ options: -march=native

Google Draco

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Lioncbaa16003200480064008000SE +/- 1.86, N = 37332732073511. (CXX) g++ options: -O3
OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Lioncbaa13002600390052006500Min: 7349 / Avg: 7351.33 / Max: 73551. (CXX) g++ options: -O3

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa1224364860SE +/- 0.11, N = 355.2455.2655.03
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa1122334455Min: 54.84 / Avg: 55.03 / Max: 55.21

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa2004006008001000SE +/- 2.84, N = 31144.771144.801149.47
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa2004006008001000Min: 1144.1 / Avg: 1149.47 / Max: 1153.78

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamcbaa714212835SE +/- 0.01, N = 330.6730.7230.60
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamcbaa714212835Min: 30.58 / Avg: 30.6 / Max: 30.61

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamcbaa816243240SE +/- 0.01, N = 332.5832.5332.66
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamcbaa714212835Min: 32.65 / Avg: 32.66 / Max: 32.67

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamcbaa816243240SE +/- 0.04, N = 333.6733.5833.53
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamcbaa714212835Min: 33.49 / Avg: 33.53 / Max: 33.61

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamcbaa306090120150SE +/- 0.11, N = 3133.49134.00133.53
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamcbaa306090120150Min: 133.3 / Avg: 133.53 / Max: 133.66

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamcbaa246810SE +/- 0.0064, N = 37.47677.44847.4741
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamcbaa3691215Min: 7.47 / Avg: 7.47 / Max: 7.49

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUcbaa0.61881.23761.85642.47523.094SE +/- 0.00, N = 32.752.752.741. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUcbaa246810Min: 2.74 / Avg: 2.74 / Max: 2.751. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Primesieve

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13cbaa1020304050SE +/- 0.07, N = 342.2942.4442.311. (CXX) g++ options: -O3
OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13cbaa918273645Min: 42.2 / Avg: 42.3 / Max: 42.441. (CXX) g++ options: -O3

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Convolution Batch Shapes Auto - Engine: CPUcbaa0.96631.93262.89893.86524.8315SE +/- 0.01638, N = 34.284614.280364.29470MIN: 4.14MIN: 4.17MIN: 4.161. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Convolution Batch Shapes Auto - Engine: CPUcbaa246810Min: 4.27 / Avg: 4.29 / Max: 4.321. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUcbaa5001000150020002500SE +/- 1.19, N = 32157.452151.852150.30MIN: 644.54 / MAX: 2962.51MIN: 500.93 / MAX: 2975.2MIN: 491.1 / MAX: 2996.721. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUcbaa400800120016002000Min: 2148.17 / Avg: 2150.3 / Max: 2152.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 3D - Engine: CPUcbaa0.48510.97021.45531.94042.4255SE +/- 0.00137, N = 32.148782.151782.15582MIN: 2.06MIN: 2.06MIN: 2.061. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: IP Shapes 3D - Engine: CPUcbaa246810Min: 2.15 / Avg: 2.16 / Max: 2.161. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUcbaa2K4K6K8K10KSE +/- 9.32, N = 311196.5411206.1311232.43MIN: 7222.84 / MAX: 20603.63MIN: 7011.32 / MAX: 20429.17MIN: 6926.76 / MAX: 21113.441. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16-INT8 - Device: CPUcbaa2K4K6K8K10KMin: 11218.49 / Avg: 11232.43 / Max: 11250.121. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 4Kcbaaa612182430SE +/- 0.01, N = 324.9525.0124.9324.951. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 8 - Input: Bosphorus 4Kcbaaa612182430Min: 24.91 / Avg: 24.93 / Max: 24.951. (CXX) g++ options: -march=native

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUcbaa816243240SE +/- 0.02, N = 334.8834.7934.901. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUcbaa714212835Min: 34.87 / Avg: 34.9 / Max: 34.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa100200300400500SE +/- 0.42, N = 3438.25439.60438.71
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa80160240320400Min: 438.22 / Avg: 438.71 / Max: 439.55

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUcbaa80160240320400SE +/- 0.11, N = 3357.41358.50357.86MIN: 204.13 / MAX: 519.56MIN: 300.19 / MAX: 528.83MIN: 301.59 / MAX: 522.851. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUcbaa60120180240300Min: 357.69 / Avg: 357.86 / Max: 358.061. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 4Kcbaaa0.59671.19341.79012.38682.9835SE +/- 0.004, N = 32.6502.6502.6442.6521. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 4Kcbaaa246810Min: 2.64 / Avg: 2.64 / Max: 2.651. (CXX) g++ options: -march=native

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamcbaa30060090012001500SE +/- 3.19, N = 31333.721335.351337.59
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamcbaa2004006008001000Min: 1331.26 / Avg: 1337.59 / Max: 1341.49

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa306090120150SE +/- 0.16, N = 3112.85112.83112.53
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa20406080100Min: 112.25 / Avg: 112.53 / Max: 112.82

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa246810SE +/- 0.0129, N = 38.84618.84838.8709
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa3691215Min: 8.85 / Avg: 8.87 / Max: 8.89

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamcbaa400800120016002000SE +/- 1.00, N = 31836.791835.261840.37
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamcbaa30060090012001500Min: 1838.41 / Avg: 1840.37 / Max: 1841.72

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUcbaa48121620SE +/- 0.01, N = 314.7314.7714.771. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Detection FP16 - Device: CPUcbaa48121620Min: 14.76 / Avg: 14.77 / Max: 14.791. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa1122334455SE +/- 0.08, N = 350.6550.7350.60
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa1020304050Min: 50.47 / Avg: 50.6 / Max: 50.74

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa510152025SE +/- 0.03, N = 319.7319.6919.75
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamcbaa510152025Min: 19.69 / Avg: 19.75 / Max: 19.8

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUcbaa2004006008001000SE +/- 0.93, N = 3792.00793.31794.06MIN: 568.74 / MAX: 1657.2MIN: 559.01 / MAX: 1581.54MIN: 604.52 / MAX: 1620.51. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUcbaa140280420560700Min: 793.11 / Avg: 794.06 / Max: 795.931. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUcbaa918273645SE +/- 0.05, N = 340.2140.1540.111. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Machine Translation EN To DE FP16 - Device: CPUcbaa816243240Min: 40.01 / Avg: 40.11 / Max: 40.161. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa306090120150SE +/- 0.07, N = 3143.60143.48143.83
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamcbaa306090120150Min: 143.7 / Avg: 143.83 / Max: 143.96

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamcbaa1122334455SE +/- 0.11, N = 346.6846.7246.61
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamcbaa1020304050Min: 46.47 / Avg: 46.61 / Max: 46.83

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUcbaa2004006008001000SE +/- 0.49, N = 3913.21915.17913.41MIN: 718.49 / MAX: 1350.67MIN: 711.5 / MAX: 1350.07MIN: 742.17 / MAX: 1356.421. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16-INT8 - Device: CPUcbaa160320480640800Min: 912.89 / Avg: 913.41 / Max: 914.381. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamcbaa246810SE +/- 0.0095, N = 37.45407.46927.4691
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamcbaa3691215Min: 7.46 / Avg: 7.47 / Max: 7.49

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamcbaa306090120150SE +/- 0.17, N = 3133.89133.63133.62
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.7Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamcbaa306090120150Min: 133.28 / Avg: 133.62 / Max: 133.8

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUcbaa20406080100SE +/- 0.03, N = 389.3089.1989.351. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Vehicle Detection FP16-INT8 - Device: CPUcbaa20406080100Min: 89.3 / Avg: 89.35 / Max: 89.391. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPUcbaa1530456075SE +/- 0.12, N = 365.6065.4965.601. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPUcbaa1326395265Min: 65.39 / Avg: 65.6 / Max: 65.81. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPUcbaa110220330440550SE +/- 0.88, N = 3486.11486.90486.11MIN: 171.7 / MAX: 813.73MIN: 119.18 / MAX: 852.49MIN: 118.22 / MAX: 849.311. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Road Segmentation ADAS FP16 - Device: CPUcbaa90180270360450Min: 484.59 / Avg: 486.11 / Max: 487.651. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPUcbaa2K4K6K8K10KSE +/- 17.40, N = 310876.7010891.9310877.53MIN: 3255.92 / MAX: 18738.42MIN: 3821.31 / MAX: 19031.99MIN: 4104.89 / MAX: 18949.051. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPUcbaa2K4K6K8K10KMin: 10842.87 / Avg: 10877.53 / Max: 10897.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 1080pcbaaa246810SE +/- 0.010, N = 38.9268.9218.9258.9141. (CXX) g++ options: -march=native
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.0Encoder Mode: Preset 4 - Input: Bosphorus 1080pcbaaa3691215Min: 8.91 / Avg: 8.93 / Max: 8.941. (CXX) g++ options: -march=native

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPUcbaa4080120160200SE +/- 0.08, N = 3194.65194.84194.88MIN: 185.45 / MAX: 358.03MIN: 185.09 / MAX: 355.83MIN: 185.7 / MAX: 356.131. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPUcbaa4080120160200Min: 194.76 / Avg: 194.88 / Max: 195.021. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPUcbaa4080120160200SE +/- 0.06, N = 3164.13163.98163.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Handwritten English Recognition FP16 - Device: CPUcbaa306090120150Min: 163.85 / Avg: 163.95 / Max: 164.061. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUcbaa510152025SE +/- 0.05, N = 322.7822.7922.80MIN: 1.63 / MAX: 162.11MIN: 1.59 / MAX: 165.35MIN: 1.57 / MAX: 164.421. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUcbaa510152025Min: 22.71 / Avg: 22.8 / Max: 22.871. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUcbaa30060090012001500SE +/- 3.07, N = 31403.651402.971402.511. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUcbaa2004006008001000Min: 1398.45 / Avg: 1402.51 / Max: 1408.521. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

srsRAN Project

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PDSCH Processor Benchmark, Throughput Threadaaa4080120160200SE +/- 0.03, N = 3175.7175.81. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl
OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PDSCH Processor Benchmark, Throughput Threadaaa306090120150Min: 175.6 / Avg: 175.67 / Max: 175.71. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenVINO

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUcbaa4080120160200SE +/- 0.04, N = 3193.87193.93193.84MIN: 182.85 / MAX: 406.51MIN: 182.93 / MAX: 402.18MIN: 183.19 / MAX: 407.141. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUcbaa4080120160200Min: 193.78 / Avg: 193.84 / Max: 193.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUcbaa4080120160200SE +/- 0.03, N = 3164.79164.75164.821. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Noise Suppression Poconet-Like FP16 - Device: CPUcbaa306090120150Min: 164.78 / Avg: 164.82 / Max: 164.871. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUcbaa306090120150SE +/- 0.34, N = 3142.54142.58142.601. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUcbaa306090120150Min: 142.1 / Avg: 142.6 / Max: 143.251. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUcbaa50100150200250SE +/- 0.53, N = 3224.31224.22224.22MIN: 31.77 / MAX: 351.21MIN: 36.4 / MAX: 368.76MIN: 29.21 / MAX: 400.611. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2024.0Model: Person Re-Identification Retail FP16 - Device: CPUcbaa4080120160200Min: 223.22 / Avg: 224.22 / Max: 2251. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Training - Engine: CPUcbaa8001600240032004000SE +/- 2.30, N = 33738.533737.153738.39MIN: 3730.99MIN: 3730.87MIN: 3728.791. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.4Harness: Recurrent Neural Network Training - Engine: CPUcbaa6001200180024003000Min: 3735.94 / Avg: 3738.39 / Max: 3742.991. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=generic -fPIC -pie -ldl -lpthread

WavPack Audio Encoding

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.7WAV To WavPackcbaa612182430SE +/- 0.00, N = 525.2025.2125.20
OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.7WAV To WavPackcbaa612182430Min: 25.19 / Avg: 25.2 / Max: 25.21

OpenVINO

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPUcbaa0.6391.2781.9172.5563.195SE +/- 0.01, N = 32.842.842.841. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2024.0Model: Face Detection FP16 - Device: CPUcbaa246810Min: 2.83 / Avg: 2.84 / Max: 2.851. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

srsRAN Project

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PUSCH Processor Benchmark, Throughput Threada112233445546.7MIN: 28.91. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240219Test: PUSCH Processor Benchmark, Throughput Totala300600900120015001602.1MIN: 947.21. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

Stockfish

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess Benchmarkcbaaa13M26M39M52M65MSE +/- 1497045.19, N = 12535149965190185359449725590287751. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess Benchmarkcbaaa10M20M30M40M50MMin: 52770867 / Avg: 59449724.75 / Max: 691258641. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver

120 Results Shown

JPEG-XL Decoding libjxl
JPEG-XL libjxl:
  PNG - 80
  JPEG - 90
  PNG - 90
  JPEG - 80
Google Draco
oneDNN
Neural Magic DeepSparse:
  ResNet-50, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Timed Linux Kernel Compilation
OpenVINO
JPEG-XL libjxl
OpenVINO:
  Face Detection Retail FP16 - CPU
  Weld Porosity Detection FP16-INT8 - CPU
  Weld Porosity Detection FP16-INT8 - CPU
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    ms/batch
    items/sec
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    items/sec
    ms/batch
OpenVINO:
  Weld Porosity Detection FP16 - CPU:
    ms
    FPS
Primesieve
JPEG-XL libjxl
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    FPS
    ms
Neural Magic DeepSparse:
  ResNet-50, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
srsRAN Project
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
  ResNet-50, Baseline - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Parallel BZIP2 Compression
Neural Magic DeepSparse
oneDNN
JPEG-XL Decoding libjxl
OpenVINO:
  Face Detection Retail FP16-INT8 - CPU:
    FPS
    ms
SVT-AV1
Neural Magic DeepSparse
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    FPS
    ms
Neural Magic DeepSparse
OpenVINO
Neural Magic DeepSparse
oneDNN
OpenVINO
Neural Magic DeepSparse
Timed Linux Kernel Compilation
SVT-AV1
oneDNN
OpenVINO
Neural Magic DeepSparse
OpenVINO
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    items/sec
    ms/batch
SVT-AV1
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream
  Llama2 Chat 7b Quantized - Synchronous Single-Stream
  Llama2 Chat 7b Quantized - Synchronous Single-Stream
OpenVINO
SVT-AV1
OpenVINO
Neural Magic DeepSparse
SVT-AV1
Google Draco
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream:
    items/sec
    ms/batch
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    items/sec
  ResNet-50, Baseline - Synchronous Single-Stream:
    items/sec
    ms/batch
OpenVINO
Primesieve
oneDNN
OpenVINO
oneDNN
OpenVINO
SVT-AV1
OpenVINO
Neural Magic DeepSparse
OpenVINO
SVT-AV1
Neural Magic DeepSparse:
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream
  CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream
  CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
OpenVINO
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
OpenVINO:
  Machine Translation EN To DE FP16 - CPU:
    ms
    FPS
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream
OpenVINO
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenVINO:
  Vehicle Detection FP16-INT8 - CPU
  Road Segmentation ADAS FP16 - CPU
  Road Segmentation ADAS FP16 - CPU
  Face Detection FP16 - CPU
SVT-AV1
OpenVINO:
  Handwritten English Recognition FP16 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP16 - CPU:
    ms
    FPS
srsRAN Project
OpenVINO:
  Noise Suppression Poconet-Like FP16 - CPU:
    ms
    FPS
  Person Re-Identification Retail FP16 - CPU:
    FPS
    ms
oneDNN
WavPack Audio Encoding
OpenVINO
srsRAN Project:
  PUSCH Processor Benchmark, Throughput Thread
  PUSCH Processor Benchmark, Throughput Total
Stockfish

a

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 March 2024 01:23 by user root.

aa

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 March 2024 01:47 by user root.

b

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 March 2024 09:55 by user root.

c

Processor: ARMv8 Neoverse-N1 @ 3.00GHz (128 Cores), Motherboard: GIGABYTE G242-P36-00 MP32-AR2-00 v01000100 (F31k SCP: 2.10.20220531 BIOS), Chipset: Ampere Computing LLC Altra PCI Root Complex A, Memory: 16 x 32 GB DDR4-3200MT/s Samsung M393A4K40DB3-CWE, Disk: 800GB Micron_7450_MTFDKBA800TFS, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Intel I350

OS: Ubuntu 23.10, Kernel: 6.5.0-15-generic (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1024x768

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.11.6
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 March 2024 13:52 by user root.