epyc last

AMD EPYC 7343 16-Core testing with a Supermicro H12SSL-i v1.02 (2.4 BIOS) and astdrmfb on AlmaLinux 9.1 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2304307-NE-EPYCLAST283&rdt&grr.

epyc lastProcessorMotherboardMemoryDiskGraphicsMonitorOSKernelCompilerFile-SystemScreen ResolutionabcdAMD EPYC 7343 16-Core @ 3.20GHz (16 Cores / 32 Threads)Supermicro H12SSL-i v1.02 (2.4 BIOS)8 x 64 GB DDR4-3200MT/s Samsung M393A8G40AB2-CWE2 x 1920GB SAMSUNG MZQL21T9HCJR-00A07astdrmfbDELL E207WFPAlmaLinux 9.15.14.0-162.12.1.el9_1.x86_64 (x86_64)GCC 11.3.1 20220421ext41680x1050OpenBenchmarking.orgKernel Details- Transparent Huge Pages: alwaysCompiler Details- --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-host-bind-now --enable-host-pie --enable-initfini-array --enable-languages=c,c++,fortran,lto --enable-link-serialization=1 --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-arch_64=x86-64-v2 --with-build-config=bootstrap-lto --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver --without-isl Disk Details- NONE / relatime,rw,stripe=32 / raid1 nvme1n1p3[0] nvme0n1p3[1] Block Size: 4096Processor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173 Python Details- Python 3.9.14Security Details- 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 Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

epyc lastintel-tensorflow: mobilenetv1_fp32_pretrained_model - 960intel-tensorflow: inceptionv4_fp32_pretrained_model - 960intel-tensorflow: mobilenetv1_fp32_pretrained_model - 512intel-tensorflow: mobilenetv1_int8_pretrained_model - 960intel-tensorflow: inceptionv4_fp32_pretrained_model - 512intel-tensorflow: inceptionv4_int8_pretrained_model - 960intel-tensorflow: mobilenetv1_fp32_pretrained_model - 256intel-tensorflow: mobilenetv1_int8_pretrained_model - 512intel-tensorflow: resnet50_fp32_pretrained_model - 960intel-tensorflow: inceptionv4_fp32_pretrained_model - 256intel-tensorflow: inceptionv4_int8_pretrained_model - 512intel-tensorflow: resnet50_fp32_pretrained_model - 512intel-tensorflow: mobilenetv1_int8_pretrained_model - 256intel-tensorflow: mobilenetv1_fp32_pretrained_model - 96intel-tensorflow: inceptionv4_int8_pretrained_model - 256intel-tensorflow: resnet50_int8_pretrained_model - 960intel-tensorflow: resnet50_fp32_pretrained_model - 256intel-tensorflow: inceptionv4_fp32_pretrained_model - 96intel-tensorflow: mobilenetv1_fp32_pretrained_model - 64intel-tensorflow: inceptionv4_fp32_pretrained_model - 64intel-tensorflow: resnet50_fp32_pretrained_model - 96intel-tensorflow: resnet50_int8_pretrained_model - 512intel-tensorflow: mobilenetv1_int8_pretrained_model - 96influxdb: 4 - 10000 - 2,5000,1 - 10000influxdb: 64 - 10000 - 2,5000,1 - 10000intel-tensorflow: resnet50_fp32_pretrained_model - 64intel-tensorflow: inceptionv4_int8_pretrained_model - 96intel-tensorflow: resnet50_fp32_pretrained_model - 32intel-tensorflow: mobilenetv1_fp32_pretrained_model - 32intel-tensorflow: resnet50_fp32_pretrained_model - 16intel-tensorflow: mobilenetv1_int8_pretrained_model - 64intel-tensorflow: resnet50_fp32_pretrained_model - 1intel-tensorflow: resnet50_fp32_pretrained_model - 1svt-av1: Preset 4 - Bosphorus 4Kintel-tensorflow: mobilenetv1_int8_pretrained_model - 32intel-tensorflow: inceptionv4_fp32_pretrained_model - 32intel-tensorflow: resnet50_int8_pretrained_model - 256intel-tensorflow: inceptionv4_int8_pretrained_model - 64intel-tensorflow: mobilenetv1_fp32_pretrained_model - 16quantlib: intel-tensorflow: inceptionv4_int8_pretrained_model - 32sqlite: 16intel-tensorflow: inceptionv4_fp32_pretrained_model - 16svt-av1: Preset 4 - Bosphorus 1080pintel-tensorflow: resnet50_int8_pretrained_model - 96intel-tensorflow: inceptionv4_int8_pretrained_model - 16intel-tensorflow: mobilenetv1_int8_pretrained_model - 16intel-tensorflow: resnet50_int8_pretrained_model - 64svt-av1: Preset 8 - Bosphorus 4Ksqlite: 32intel-tensorflow: resnet50_int8_pretrained_model - 32sqlite: 4intel-tensorflow: mobilenetv1_fp32_pretrained_model - 1intel-tensorflow: resnet50_int8_pretrained_model - 16intel-tensorflow: inceptionv4_int8_pretrained_model - 1intel-tensorflow: inceptionv4_int8_pretrained_model - 1svt-av1: Preset 8 - Bosphorus 1080pintel-tensorflow: resnet50_int8_pretrained_model - 1intel-tensorflow: resnet50_int8_pretrained_model - 1intel-tensorflow: inceptionv4_fp32_pretrained_model - 1intel-tensorflow: inceptionv4_fp32_pretrained_model - 1intel-tensorflow: mobilenetv1_int8_pretrained_model - 1sqlite: 8svt-av1: Preset 13 - Bosphorus 4Ksvt-av1: Preset 12 - Bosphorus 4Ksqlite: 2svt-av1: Preset 12 - Bosphorus 1080psvt-av1: Preset 13 - Bosphorus 1080pabcd983.7651.76976.582133.1851.76120.741001.612170.09168.72151.92119.90168.3712090.97990.35119.18391.680167.96851.83998.4352.44169.726383.6152083.551547894.41602099.9170.509118.25174.040981.32168.7442112.3312.61379.2833.7662056.1853.10382.093118.48932.193202.1117.008.47653.209.027373.126113.312003.49365.09552.58211.321356.3173.1781045.59346.01714.43669.0495.9254.508221.85530.81732.231933.375.065160.501174.5192.150547.498548.007982.7151.78974.842128.151.87120.941000.282179.09169.72952.07119.61168.6412106.54986.72119.18392.07167.99851.72997.7352.40169.649385.5462081.871545780.81593776.4170.516119.097759734174.303984.41171.6322120.7712.53579.7763.7822110.1952.91380.652118.93929.583206.1117.816.20953.479.064373.723111.632002.09364.04351.98211.811361.3622.9381048.2347.92614.41569.1695.7074.509221.76930.47533.161932.473.856160.664175.6832.041542.028542.625982.4651.59976.222137.251.77120.741001.432161.98170.09352.06119.91168.7882028.82988.15119.33392.383168.05551.90999.9351.72169.971385.52087.391552035.61599391.9170.603119.82174.255982.66171.0492063.7812.50279.9883.7912037.8653.11381.064117.61933.763200.7116.937.19852.999.121372.933113.631989365.00352.52311.743357.7142.9041046.79348.36514.42669.0196.3294.622216.36930.64732.061933.583.966160.139172.6962.106535.678545.864984.3651.62974.692132.2951.76120.571001.32171.79169.33852.04120.56168.3652091.6988.72119.45391.383168.1652.00999.9251.97169.31384.262071.1815607581600346.9170.261118.23172.269981.61169.3332091.6612.46680.2183.7842033.5753.27380.015119.83931.223192.7117.896.07553.229.28372.928113.361984.38365.30552.57211.598357.0352.7121046.4344.86814.37769.0795.6484.596217.59330.72331.861934.323.761159.521174.8362.039539.588547.432OpenBenchmarking.org

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 960abcd2004006008001000SE +/- 0.13, N = 3983.76982.71982.46984.36

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 960abcd1224364860SE +/- 0.11, N = 351.7651.7851.5951.62

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 512abcd2004006008001000SE +/- 0.75, N = 3976.58974.84976.22974.69

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 960abcd5001000150020002500SE +/- 2.58, N = 32133.182128.102137.202132.29

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 512abcd1224364860SE +/- 0.05, N = 351.7651.8751.7751.76

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 960abcd306090120150SE +/- 0.16, N = 3120.74120.94120.74120.57

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 256abcd2004006008001000SE +/- 0.11, N = 31001.611000.281001.431001.30

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 512abcd5001000150020002500SE +/- 2.76, N = 32170.092179.092161.982171.79

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 960abcd4080120160200SE +/- 0.30, N = 3168.72169.73170.09169.34

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 256abcd1224364860SE +/- 0.02, N = 351.9252.0752.0652.04

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 512abcd306090120150SE +/- 0.14, N = 3119.90119.61119.91120.56

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 512abcd4080120160200SE +/- 0.21, N = 3168.37168.64168.79168.37

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 256abcd5001000150020002500SE +/- 14.60, N = 32090.972106.542028.822091.60

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 96abcd2004006008001000SE +/- 1.26, N = 3990.35986.72988.15988.72

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 256abcd306090120150SE +/- 0.27, N = 3119.18119.18119.33119.45

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 960

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 960abcd90180270360450SE +/- 0.57, N = 3391.68392.07392.38391.38

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 256abcd4080120160200SE +/- 0.12, N = 3167.97168.00168.06168.16

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 96abcd1224364860SE +/- 0.07, N = 351.8351.7251.9052.00

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 64abcd2004006008001000SE +/- 0.55, N = 3998.43997.73999.93999.92

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 64abcd1224364860SE +/- 0.12, N = 352.4452.4051.7251.97

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 96abcd4080120160200SE +/- 0.11, N = 3169.73169.65169.97169.31

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 512

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 512abcd80160240320400SE +/- 0.21, N = 3383.62385.55385.50384.26

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 96abcd400800120016002000SE +/- 2.02, N = 32083.552081.872087.392071.18

InfluxDB

Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000abcd300K600K900K1200K1500KSE +/- 5338.36, N = 31547894.41545780.81552035.61560758.0

InfluxDB

Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000abcd300K600K900K1200K1500KSE +/- 3918.66, N = 31602099.91593776.41599391.91600346.9

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 64abcd4080120160200SE +/- 0.12, N = 3170.51170.52170.60170.26

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 96abcd306090120150SE +/- 0.42, N = 3118.25119.10119.82118.23

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 32abcd4080120160200SE +/- 0.31, N = 3174.04174.30174.26172.27

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 32abcd2004006008001000SE +/- 1.14, N = 3981.32984.41982.66981.61

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 16abcd4080120160200SE +/- 0.83, N = 3168.74171.63171.05169.33

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 64abcd5001000150020002500SE +/- 10.67, N = 32112.332120.772063.782091.66

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 1

OpenBenchmarking.orgms, Fewer Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 1abcd3691215SE +/- 0.01, N = 312.6112.5412.5012.47

Intel TensorFlow

Model: resnet50_fp32_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_fp32_pretrained_model - Batch Size: 1abcd20406080100SE +/- 0.09, N = 379.2879.7879.9980.22

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 4 - Input: Bosphorus 4Kabcd0.8531.7062.5593.4124.265SE +/- 0.019, N = 33.7663.7823.7913.7841. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 32abcd5001000150020002500SE +/- 19.33, N = 72056.182110.192037.862033.57

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 32abcd1224364860SE +/- 0.09, N = 353.1052.9153.1153.27

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 256

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 256abcd80160240320400SE +/- 0.54, N = 3382.09380.65381.06380.02

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 64abcd306090120150SE +/- 0.54, N = 3118.48118.93117.61119.83

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 16abcd2004006008001000SE +/- 0.39, N = 3932.19929.58933.76931.22

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.30abcd7001400210028003500SE +/- 1.35, N = 33202.13206.13200.73192.71. (CXX) g++ options: -O3 -march=native -fPIE -pie

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 32abcd306090120150SE +/- 0.81, N = 3117.00117.81116.93117.89

SQLite

Threads / Copies: 16

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 16abcd246810SE +/- 0.163, N = 138.4766.2097.1986.0751. (CC) gcc options: -O2 -lz -lm

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 16abcd1224364860SE +/- 0.07, N = 353.2053.4752.9953.22

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 4 - Input: Bosphorus 1080pabcd3691215SE +/- 0.031, N = 39.0279.0649.1219.2801. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 96

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 96abcd80160240320400SE +/- 0.27, N = 3373.13373.72372.93372.93

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 16abcd306090120150SE +/- 0.30, N = 3113.31111.63113.63113.36

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 16abcd400800120016002000SE +/- 14.27, N = 32003.492002.091989.001984.38

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 64

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 64abcd80160240320400SE +/- 0.43, N = 3365.10364.04365.00365.31

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 8 - Input: Bosphorus 4Kabcd1224364860SE +/- 0.19, N = 352.5851.9852.5252.571. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SQLite

Threads / Copies: 32

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 32abcd3691215SE +/- 0.05, N = 311.3211.8111.7411.601. (CC) gcc options: -O2 -lz -lm

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 32

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 32abcd80160240320400SE +/- 0.47, N = 3356.32361.36357.71357.04

SQLite

Threads / Copies: 4

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 4abcd0.71511.43022.14532.86043.5755SE +/- 0.030, N = 153.1782.9382.9042.7121. (CC) gcc options: -O2 -lz -lm

Intel TensorFlow

Model: mobilenetv1_fp32_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_fp32_pretrained_model - Batch Size: 1abcd2004006008001000SE +/- 0.98, N = 31045.591048.201046.791046.40

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 16

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 16abcd80160240320400SE +/- 1.45, N = 3346.02347.93348.37344.87

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 1

OpenBenchmarking.orgms, Fewer Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 1abcd48121620SE +/- 0.03, N = 314.4414.4214.4314.38

Intel TensorFlow

Model: inceptionv4_int8_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_int8_pretrained_model - Batch Size: 1abcd1530456075SE +/- 0.07, N = 369.0469.1669.0169.07

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 8 - Input: Bosphorus 1080pabcd20406080100SE +/- 0.42, N = 395.9395.7196.3395.651. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 1

OpenBenchmarking.orgms, Fewer Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 1abcd1.042.083.124.165.2SE +/- 0.032, N = 34.5084.5094.6224.596

Intel TensorFlow

Model: resnet50_int8_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: resnet50_int8_pretrained_model - Batch Size: 1abcd50100150200250SE +/- 1.58, N = 3221.86221.77216.37217.59

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 1

OpenBenchmarking.orgms, Fewer Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 1abcd714212835SE +/- 0.10, N = 330.8230.4830.6530.72

Intel TensorFlow

Model: inceptionv4_fp32_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: inceptionv4_fp32_pretrained_model - Batch Size: 1abcd816243240SE +/- 0.26, N = 332.2333.1632.0631.86

Intel TensorFlow

Model: mobilenetv1_int8_pretrained_model - Batch Size: 1

OpenBenchmarking.orgimages/sec, More Is BetterIntel TensorFlow 2.12Model: mobilenetv1_int8_pretrained_model - Batch Size: 1abcd400800120016002000SE +/- 0.61, N = 31933.371932.471933.581934.32

SQLite

Threads / Copies: 8

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 8abcd1.13962.27923.41884.55845.698SE +/- 0.038, N = 35.0653.8563.9663.7611. (CC) gcc options: -O2 -lz -lm

SVT-AV1

Encoder Mode: Preset 13 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 13 - Input: Bosphorus 4Kabcd4080120160200SE +/- 0.85, N = 3160.50160.66160.14159.521. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 12 - Input: Bosphorus 4Kabcd4080120160200SE +/- 0.56, N = 3174.52175.68172.70174.841. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SQLite

Threads / Copies: 2

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 2abcd0.48380.96761.45141.93522.419SE +/- 0.004, N = 32.1502.0412.1062.0391. (CC) gcc options: -O2 -lz -lm

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 12 - Input: Bosphorus 1080pabcd120240360480600SE +/- 0.64, N = 3547.50542.03535.68539.591. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SVT-AV1

Encoder Mode: Preset 13 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.5Encoder Mode: Preset 13 - Input: Bosphorus 1080pabcd120240360480600SE +/- 0.34, N = 3548.01542.63545.86547.431. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Phoronix Test Suite v10.8.5