eps Tests for a future article. 2 x AMD EPYC 9684X 96-Core testing with a AMD Titanite_4G (RTI1007B BIOS) and ASPEED on Ubuntu 23.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2312240-NE-EPS17737430&grs&sor .
eps Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Compiler File-System Screen Resolution a b 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads) AMD Titanite_4G (RTI1007B BIOS) AMD Device 14a4 1520GB 3201GB Micron_7450_MTFDKCB3T2TFS ASPEED Broadcom NetXtreme BCM5720 PCIe Ubuntu 23.10 6.5.0-13-generic (x86_64) GCC 13.2.0 ext4 800x600 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e Java Details - OpenJDK Runtime Environment (build 11.0.21+9-post-Ubuntu-0ubuntu123.10) Python Details - Python 3.11.6 Security Details - 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: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
eps pytorch: CPU - 256 - ResNet-152 pytorch: CPU - 1 - Efficientnet_v2_l svt-av1: Preset 13 - Bosphorus 4K svt-av1: Preset 12 - Bosphorus 4K webp2: Quality 100, Compression Effort 5 pytorch: CPU - 256 - ResNet-50 pytorch: CPU - 1 - ResNet-152 pytorch: CPU - 1 - ResNet-50 pytorch: CPU - 16 - ResNet-50 spark-tpch: 1 - Geometric Mean Of All Queries pytorch: CPU - 256 - Efficientnet_v2_l webp2: Default svt-av1: Preset 8 - Bosphorus 1080p deepsparse: ResNet-50, Baseline - Synchronous Single-Stream deepsparse: ResNet-50, Baseline - Synchronous Single-Stream webp2: Quality 75, Compression Effort 7 xmrig: CryptoNight-Femto UPX2 - 1M pytorch: CPU - 32 - ResNet-152 pytorch: CPU - 16 - Efficientnet_v2_l java-scimark2: Sparse Matrix Multiply xmrig: CryptoNight-Heavy - 1M spark-tpch: 10 - Geometric Mean Of All Queries java-scimark2: Dense LU Matrix Factorization svt-av1: Preset 4 - Bosphorus 1080p deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream svt-av1: Preset 13 - Bosphorus 1080p svt-av1: Preset 4 - Bosphorus 4K svt-av1: Preset 8 - Bosphorus 4K deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream pytorch: CPU - 16 - ResNet-152 deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream pytorch: CPU - 32 - ResNet-50 xmrig: GhostRider - 1M deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream xmrig: Wownero - 1M deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream svt-av1: Preset 12 - Bosphorus 1080p deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream xmrig: Monero - 1M deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream java-scimark2: Composite deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream java-scimark2: Fast Fourier Transform deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream openssl: SHA512 deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream openssl: SHA256 xmrig: KawPow - 1M spark-tpch: 50 - Geometric Mean Of All Queries deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream openssl: RSA4096 deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Stream deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Stream java-scimark2: Monte Carlo openssl: RSA4096 deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream java-scimark2: Jacobi Successive Over-Relaxation deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream pytorch: CPU - 32 - Efficientnet_v2_l webp2: Quality 100, Lossless Compression webp2: Quality 95, Compression Effort 7 spark-tpch: 50 - Q22 spark-tpch: 50 - Q21 spark-tpch: 50 - Q20 spark-tpch: 50 - Q19 spark-tpch: 50 - Q18 spark-tpch: 50 - Q17 spark-tpch: 50 - Q16 spark-tpch: 50 - Q15 spark-tpch: 50 - Q14 spark-tpch: 50 - Q13 spark-tpch: 50 - Q12 spark-tpch: 50 - Q11 spark-tpch: 50 - Q10 spark-tpch: 50 - Q09 spark-tpch: 50 - Q08 spark-tpch: 50 - Q07 spark-tpch: 50 - Q06 spark-tpch: 50 - Q05 spark-tpch: 50 - Q04 spark-tpch: 50 - Q03 spark-tpch: 50 - Q02 spark-tpch: 50 - Q01 spark-tpch: 10 - Q22 spark-tpch: 10 - Q21 spark-tpch: 10 - Q20 spark-tpch: 10 - Q19 spark-tpch: 10 - Q18 spark-tpch: 10 - Q17 spark-tpch: 10 - Q16 spark-tpch: 10 - Q15 spark-tpch: 10 - Q14 spark-tpch: 10 - Q13 spark-tpch: 10 - Q12 spark-tpch: 10 - Q11 spark-tpch: 10 - Q10 spark-tpch: 10 - Q09 spark-tpch: 10 - Q08 spark-tpch: 10 - Q07 spark-tpch: 10 - Q06 spark-tpch: 10 - Q05 spark-tpch: 10 - Q04 spark-tpch: 10 - Q03 spark-tpch: 10 - Q02 spark-tpch: 10 - Q01 spark-tpch: 1 - Q22 spark-tpch: 1 - Q21 spark-tpch: 1 - Q20 spark-tpch: 1 - Q19 spark-tpch: 1 - Q18 spark-tpch: 1 - Q17 spark-tpch: 1 - Q16 spark-tpch: 1 - Q15 spark-tpch: 1 - Q14 spark-tpch: 1 - Q13 spark-tpch: 1 - Q12 spark-tpch: 1 - Q11 spark-tpch: 1 - Q10 spark-tpch: 1 - Q09 spark-tpch: 1 - Q08 spark-tpch: 1 - Q07 spark-tpch: 1 - Q06 spark-tpch: 1 - Q05 spark-tpch: 1 - Q04 spark-tpch: 1 - Q03 spark-tpch: 1 - Q02 spark-tpch: 1 - Q01 lczero: Eigen lczero: BLAS a b 8.96 6.40 176.670 178.910 6.51 21.29 10.16 23.57 21.16 2.44964916 2.32 9.48 165.104 4.7637 209.7998 0.83 123199.0 8.90 2.32 2809.01 123041.6 10.72150208 13358.53 21.424 4.8022 208.1200 635.810 8.248 86.434 2608.0090 8.93 36.7508 21.00 31859.7 248.5770 224.5798 4.4503 5.5955 131141.9 715.0362 17108.4634 122.0312 571.875 48.4917 20.6157 123352.8 15.3183 65.2070 3984.62 784.5178 132.6580 132.0485 420.74 48.4476 20.6345 91630925473 719.2814 796.0713 383.2004 190.7999 5.2377 120.2065 54.4097 32.0229 31.2154 1761.4041 281869895760 123558.6 19.58745807 1758.5931 98622.0 54.5064 1.2413 804.1784 1631.42 3244390.3 4.7126 4.7188 212.0955 211.7448 1703.42 156.4159 1136.7105 84.2519 68.2655 14.6422 17.3023 5540.6268 607.5664 2.32 0.11 0.45 10.69325638 87.89528910 20.79876137 10.45287259 34.51305643 24.30927912 14.21570397 9.77733866 12.70455011 12.75901413 19.40537771 13.58028253 24.36003748 36.66458511 26.73535283 24.85711161 5.90309207 29.83627891 20.99598312 26.18788719 14.25487200 12.00795619 6.05411895 32.90715027 11.43560823 6.20677837 18.46971194 12.77044550 6.87131294 5.84138076 7.07622369 7.37728373 9.94400438 8.00292349 15.17488098 21.90670204 15.51824761 14.65200933 2.05104745 16.44365629 12.34571203 13.97308763 7.43104283 7.58889151 1.00769047 9.64531231 3.05739617 0.79092395 5.62853845 2.95993924 1.38147259 2.50185966 2.06485331 1.58815936 2.17542648 1.27338135 3.81359665 5.70969407 2.65584644 4.01044806 0.46822915 4.13122161 3.92525745 3.86442184 2.06179071 4.32006081 704 853 9.65 6.74 184.347 186.609 6.28 20.60 10.43 23.12 21.57 2.49517747 2.28 9.63 162.561 4.829 206.969 0.82 122070.3 8.98 2.34 2792.09 123777.7 10.65793942 13434.09 21.313 4.7775 209.1955 639.088 8.208 86.841 2596.0961 8.97 36.9146 21.09 31728.9 249.4983 225.4047 4.4341 5.6159 131613.6 717.5936 17047.639 121.6067 569.955 48.332 20.6837 122971 15.3653 65.0079 3996.76 786.8905 132.2719 132.4219 421.91 48.3264 20.686 91835961470 717.9791 797.4124 382.5561 191.0971 5.2296 120.0373 54.4859 31.9795 31.2575 1759.0746 282211175400 123411.1 19.56475658 1756.5569 98528.8 54.5546 1.2404 804.7528 1632.45 3243345.2 4.7117 4.718 212.1305 211.7729 1703.25 156.4283 1136.6439 84.2491 68.2636 14.6426 17.3019 5540.517 607.5735 2.32 0.11 0.45 10.87410069 77.70675659 21.05384445 12.08592796 33.74198151 24.55788994 14.97535801 9.48287773 12.56767082 13.04496479 17.70001793 13.31200027 24.68585587 36.66526794 26.62909508 25.86055183 5.88382483 31.20059776 21.8167572 29.68590546 14.53046799 12.86835003 6.04430914 32.70154953 11.53966141 6.0604167 17.31370163 13.01374149 6.95270681 5.43870592 6.90602303 7.94083786 10.03829002 8.27814293 14.77719498 22.52552795 14.55769348 14.89605904 1.8559593 18.86129189 11.26242161 14.28984642 7.39245987 7.28826714 1.06679213 9.55909538 3.05001688 0.85797596 5.13171148 2.88348198 1.51779914 2.58714175 2.21146965 1.74074161 2.26641607 1.13998687 3.81245542 5.89775848 2.60907817 3.87790275 0.35801557 3.69217634 3.75427246 3.86610818 2.08224201 4.44657946 715 871 OpenBenchmarking.org
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 b a 3 6 9 12 15 SE +/- 0.05, N = 3 9.65 8.96 MIN: 4.98 / MAX: 9.85 MIN: 4.84 / MAX: 9.24
PyTorch Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l b a 2 4 6 8 10 SE +/- 0.05, N = 3 6.74 6.40 MIN: 3.48 / MAX: 6.89 MIN: 2.93 / MAX: 6.73
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 13 - Input: Bosphorus 4K b a 40 80 120 160 200 SE +/- 1.61, N = 15 184.35 176.67 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 12 - Input: Bosphorus 4K b a 40 80 120 160 200 SE +/- 1.43, N = 3 186.61 178.91 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
WebP2 Image Encode Encode Settings: Quality 100, Compression Effort 5 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Compression Effort 5 a b 2 4 6 8 10 SE +/- 0.04, N = 3 6.51 6.28 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 a b 5 10 15 20 25 SE +/- 0.31, N = 3 21.29 20.60 MIN: 13.22 / MAX: 22.39 MIN: 13.89 / MAX: 21.35
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 b a 3 6 9 12 15 SE +/- 0.08, N = 3 10.43 10.16 MIN: 4.8 / MAX: 11.36 MIN: 4.56 / MAX: 10.94
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 a b 6 12 18 24 30 SE +/- 0.19, N = 15 23.57 23.12 MIN: 11.38 / MAX: 25.62 MIN: 12.17 / MAX: 24.33
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: ResNet-50 b a 5 10 15 20 25 SE +/- 0.25, N = 3 21.57 21.16 MIN: 14.06 / MAX: 22.29 MIN: 12.26 / MAX: 22.24
Apache Spark TPC-H Scale Factor: 1 - Geometric Mean Of All Queries OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Geometric Mean Of All Queries a b 0.5614 1.1228 1.6842 2.2456 2.807 SE +/- 0.02040294, N = 3 2.44964916 2.49517747 MIN: 0.73 / MAX: 10.03 MIN: 0.86 / MAX: 9.56
PyTorch Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l a b 0.522 1.044 1.566 2.088 2.61 SE +/- 0.00, N = 3 2.32 2.28 MIN: 1.83 / MAX: 2.8 MIN: 1.71 / MAX: 2.84
WebP2 Image Encode Encode Settings: Default OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default b a 3 6 9 12 15 SE +/- 0.08, N = 3 9.63 9.48 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 8 - Input: Bosphorus 1080p a b 40 80 120 160 200 SE +/- 1.87, N = 3 165.10 162.56 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream a b 1.0865 2.173 3.2595 4.346 5.4325 SE +/- 0.0118, N = 3 4.7637 4.8290
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream a b 50 100 150 200 250 SE +/- 0.52, N = 3 209.80 206.97
WebP2 Image Encode Encode Settings: Quality 75, Compression Effort 7 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 a b 0.1868 0.3736 0.5604 0.7472 0.934 SE +/- 0.00, N = 3 0.83 0.82 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
Xmrig Variant: CryptoNight-Femto UPX2 - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: CryptoNight-Femto UPX2 - Hash Count: 1M a b 30K 60K 90K 120K 150K SE +/- 220.87, N = 3 123199.0 122070.3 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: ResNet-152 b a 3 6 9 12 15 SE +/- 0.10, N = 3 8.98 8.90 MIN: 5.1 / MAX: 9.29 MIN: 4.8 / MAX: 9.23
PyTorch Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l b a 0.5265 1.053 1.5795 2.106 2.6325 SE +/- 0.01, N = 3 2.34 2.32 MIN: 1.78 / MAX: 2.78 MIN: 1.77 / MAX: 2.81
Java SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Sparse Matrix Multiply a b 600 1200 1800 2400 3000 SE +/- 3.16, N = 3 2809.01 2792.09
Xmrig Variant: CryptoNight-Heavy - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: CryptoNight-Heavy - Hash Count: 1M b a 30K 60K 90K 120K 150K SE +/- 33.09, N = 3 123777.7 123041.6 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Apache Spark TPC-H Scale Factor: 10 - Geometric Mean Of All Queries OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Geometric Mean Of All Queries b a 3 6 9 12 15 SE +/- 0.02, N = 3 10.66 10.72 MIN: 5.44 / MAX: 32.7 MIN: 5.7 / MAX: 33.03
Java SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Dense LU Matrix Factorization b a 3K 6K 9K 12K 15K SE +/- 31.70, N = 3 13434.09 13358.53
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 4 - Input: Bosphorus 1080p a b 5 10 15 20 25 SE +/- 0.13, N = 3 21.42 21.31 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream b a 1.0805 2.161 3.2415 4.322 5.4025 SE +/- 0.0103, N = 3 4.7775 4.8022
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream b a 50 100 150 200 250 SE +/- 0.44, N = 3 209.20 208.12
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 13 - Input: Bosphorus 1080p b a 140 280 420 560 700 SE +/- 8.75, N = 3 639.09 635.81 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 4 - Input: Bosphorus 4K a b 2 4 6 8 10 SE +/- 0.041, N = 3 8.248 8.208 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 8 - Input: Bosphorus 4K b a 20 40 60 80 100 SE +/- 0.17, N = 3 86.84 86.43 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 600 1200 1800 2400 3000 SE +/- 6.37, N = 3 2608.01 2596.10
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: ResNet-152 b a 3 6 9 12 15 SE +/- 0.10, N = 3 8.97 8.93 MIN: 4.96 / MAX: 9.11 MIN: 4.75 / MAX: 9.39
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 8 16 24 32 40 SE +/- 0.09, N = 3 36.75 36.91
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: ResNet-50 b a 5 10 15 20 25 SE +/- 0.20, N = 3 21.09 21.00 MIN: 13.93 / MAX: 21.71 MIN: 11.39 / MAX: 21.87
Xmrig Variant: GhostRider - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: GhostRider - Hash Count: 1M a b 7K 14K 21K 28K 35K SE +/- 24.02, N = 3 31859.7 31728.9 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream b a 50 100 150 200 250 SE +/- 0.41, N = 3 249.50 248.58
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream b a 50 100 150 200 250 SE +/- 0.01, N = 3 225.40 224.58
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream b a 1.0013 2.0026 3.0039 4.0052 5.0065 SE +/- 0.0001, N = 3 4.4341 4.4503
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 1.2636 2.5272 3.7908 5.0544 6.318 SE +/- 0.0055, N = 3 5.5955 5.6159
Xmrig Variant: Wownero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: Wownero - Hash Count: 1M b a 30K 60K 90K 120K 150K SE +/- 621.69, N = 3 131613.6 131141.9 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a b 150 300 450 600 750 SE +/- 4.21, N = 3 715.04 717.59
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 4K 8K 12K 16K 20K SE +/- 16.76, N = 3 17108.46 17047.64
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream b a 30 60 90 120 150 SE +/- 0.25, N = 3 121.61 122.03
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.8 Encoder Mode: Preset 12 - Input: Bosphorus 1080p a b 120 240 360 480 600 SE +/- 1.39, N = 3 571.88 569.96 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b 11 22 33 44 55 SE +/- 0.05, N = 3 48.49 48.33
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 SE +/- 0.02, N = 3 20.62 20.68
Xmrig Variant: Monero - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: Monero - Hash Count: 1M a b 30K 60K 90K 120K 150K SE +/- 404.54, N = 3 123352.8 122971.0 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b 4 8 12 16 20 SE +/- 0.01, N = 3 15.32 15.37
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b 15 30 45 60 75 SE +/- 0.04, N = 3 65.21 65.01
Java SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Composite b a 900 1800 2700 3600 4500 SE +/- 6.24, N = 3 3996.76 3984.62
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream b a 200 400 600 800 1000 SE +/- 1.42, N = 3 786.89 784.52
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a b 30 60 90 120 150 SE +/- 0.66, N = 3 132.66 132.27
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream b a 30 60 90 120 150 SE +/- 0.03, N = 3 132.42 132.05
Java SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Fast Fourier Transform b a 90 180 270 360 450 SE +/- 0.36, N = 3 421.91 420.74
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b 11 22 33 44 55 SE +/- 0.02, N = 3 48.45 48.33
Neural Magic DeepSparse Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 SE +/- 0.01, N = 3 20.63 20.69
OpenSSL Algorithm: SHA512 OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: SHA512 b a 20000M 40000M 60000M 80000M 100000M SE +/- 191332047.54, N = 3 91835961470 91630925473 1. OpenSSL 3.0.10 1 Aug 2023 (Library: OpenSSL 3.0.10 1 Aug 2023)
Neural Magic DeepSparse Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream b a 160 320 480 640 800 SE +/- 1.53, N = 3 717.98 719.28
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 200 400 600 800 1000 SE +/- 1.54, N = 3 797.41 796.07
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream b a 80 160 240 320 400 SE +/- 0.61, N = 3 382.56 383.20
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream b a 40 80 120 160 200 SE +/- 0.06, N = 3 191.10 190.80
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream b a 1.1785 2.357 3.5355 4.714 5.8925 SE +/- 0.0015, N = 3 5.2296 5.2377
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 30 60 90 120 150 SE +/- 0.24, N = 3 120.04 120.21
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b 12 24 36 48 60 SE +/- 0.07, N = 3 54.41 54.49
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream a b 7 14 21 28 35 SE +/- 0.03, N = 3 32.02 31.98
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream a b 7 14 21 28 35 SE +/- 0.03, N = 3 31.22 31.26
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b 400 800 1200 1600 2000 SE +/- 2.24, N = 3 1761.40 1759.07
OpenSSL Algorithm: SHA256 OpenBenchmarking.org byte/s, More Is Better OpenSSL Algorithm: SHA256 b a 60000M 120000M 180000M 240000M 300000M SE +/- 548972949.20, N = 3 282211175400 281869895760 1. OpenSSL 3.0.10 1 Aug 2023 (Library: OpenSSL 3.0.10 1 Aug 2023)
Xmrig Variant: KawPow - Hash Count: 1M OpenBenchmarking.org H/s, More Is Better Xmrig 6.21 Variant: KawPow - Hash Count: 1M a b 30K 60K 90K 120K 150K SE +/- 87.00, N = 3 123558.6 123411.1 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Apache Spark TPC-H Scale Factor: 50 - Geometric Mean Of All Queries OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Geometric Mean Of All Queries b a 5 10 15 20 25 SE +/- 0.05, N = 3 19.56 19.59 MIN: 9.48 / MAX: 77.71 MIN: 9.71 / MAX: 103.64
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream a b 400 800 1200 1600 2000 SE +/- 1.91, N = 3 1758.59 1756.56
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org sign/s, More Is Better OpenSSL Algorithm: RSA4096 a b 20K 40K 60K 80K 100K SE +/- 53.45, N = 3 98622.0 98528.8 1. OpenSSL 3.0.10 1 Aug 2023 (Library: OpenSSL 3.0.10 1 Aug 2023)
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream a b 12 24 36 48 60 SE +/- 0.06, N = 3 54.51 54.55
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream b a 0.2793 0.5586 0.8379 1.1172 1.3965 SE +/- 0.0046, N = 3 1.2404 1.2413
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream b a 200 400 600 800 1000 SE +/- 3.00, N = 3 804.75 804.18
Java SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Monte Carlo b a 400 800 1200 1600 2000 SE +/- 0.75, N = 3 1632.45 1631.42
OpenSSL Algorithm: RSA4096 OpenBenchmarking.org verify/s, More Is Better OpenSSL Algorithm: RSA4096 a b 700K 1400K 2100K 2800K 3500K SE +/- 1292.47, N = 3 3244390.3 3243345.2 1. OpenSSL 3.0.10 1 Aug 2023 (Library: OpenSSL 3.0.10 1 Aug 2023)
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream b a 1.0603 2.1206 3.1809 4.2412 5.3015 SE +/- 0.0079, N = 3 4.7117 4.7126
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream b a 1.0617 2.1234 3.1851 4.2468 5.3085 SE +/- 0.0110, N = 3 4.7180 4.7188
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream b a 50 100 150 200 250 SE +/- 0.35, N = 3 212.13 212.10
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream b a 50 100 150 200 250 SE +/- 0.50, N = 3 211.77 211.74
Java SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.2 Computational Test: Jacobi Successive Over-Relaxation a b 400 800 1200 1600 2000 SE +/- 0.16, N = 3 1703.42 1703.25
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream b a 30 60 90 120 150 SE +/- 0.02, N = 3 156.43 156.42
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a b 200 400 600 800 1000 SE +/- 2.45, N = 3 1136.71 1136.64
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream b a 20 40 60 80 100 SE +/- 0.19, N = 3 84.25 84.25
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b 15 30 45 60 75 SE +/- 0.11, N = 3 68.27 68.26
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b 4 8 12 16 20 SE +/- 0.02, N = 3 14.64 14.64
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 4 8 12 16 20 SE +/- 0.01, N = 3 17.30 17.30
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 1200 2400 3600 4800 6000 SE +/- 5.02, N = 3 5540.63 5540.52
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream a b 130 260 390 520 650 SE +/- 0.37, N = 3 607.57 607.57
PyTorch Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l b a 0.522 1.044 1.566 2.088 2.61 SE +/- 0.01, N = 3 2.32 2.32 MIN: 1.93 / MAX: 2.71 MIN: 1.86 / MAX: 2.8
WebP2 Image Encode Encode Settings: Quality 100, Lossless Compression OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 100, Lossless Compression b a 0.0248 0.0496 0.0744 0.0992 0.124 SE +/- 0.00, N = 3 0.11 0.11 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
WebP2 Image Encode Encode Settings: Quality 95, Compression Effort 7 OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 b a 0.1013 0.2026 0.3039 0.4052 0.5065 SE +/- 0.00, N = 3 0.45 0.45 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
Apache Spark TPC-H Scale Factor: 50 - Q22 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q22 a b 3 6 9 12 15 SE +/- 0.12, N = 3 10.69 10.87
Apache Spark TPC-H Scale Factor: 50 - Q21 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q21 b a 20 40 60 80 100 SE +/- 7.93, N = 3 77.71 87.90
Apache Spark TPC-H Scale Factor: 50 - Q20 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q20 a b 5 10 15 20 25 SE +/- 0.10, N = 3 20.80 21.05
Apache Spark TPC-H Scale Factor: 50 - Q19 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q19 a b 3 6 9 12 15 SE +/- 0.12, N = 3 10.45 12.09
Apache Spark TPC-H Scale Factor: 50 - Q18 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q18 b a 8 16 24 32 40 SE +/- 0.33, N = 3 33.74 34.51
Apache Spark TPC-H Scale Factor: 50 - Q17 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q17 a b 6 12 18 24 30 SE +/- 0.55, N = 3 24.31 24.56
Apache Spark TPC-H Scale Factor: 50 - Q16 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q16 a b 4 8 12 16 20 SE +/- 0.26, N = 3 14.22 14.98
Apache Spark TPC-H Scale Factor: 50 - Q15 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q15 b a 3 6 9 12 15 SE +/- 0.05343306, N = 3 9.48287773 9.77733866
Apache Spark TPC-H Scale Factor: 50 - Q14 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q14 b a 3 6 9 12 15 SE +/- 0.22, N = 3 12.57 12.70
Apache Spark TPC-H Scale Factor: 50 - Q13 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q13 a b 3 6 9 12 15 SE +/- 0.08, N = 3 12.76 13.04
Apache Spark TPC-H Scale Factor: 50 - Q12 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q12 b a 5 10 15 20 25 SE +/- 1.19, N = 3 17.70 19.41
Apache Spark TPC-H Scale Factor: 50 - Q11 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q11 b a 3 6 9 12 15 SE +/- 0.32, N = 3 13.31 13.58
Apache Spark TPC-H Scale Factor: 50 - Q10 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q10 a b 6 12 18 24 30 SE +/- 0.32, N = 3 24.36 24.69
Apache Spark TPC-H Scale Factor: 50 - Q09 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q09 a b 8 16 24 32 40 SE +/- 0.34, N = 3 36.66 36.67
Apache Spark TPC-H Scale Factor: 50 - Q08 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q08 b a 6 12 18 24 30 SE +/- 0.26, N = 3 26.63 26.74
Apache Spark TPC-H Scale Factor: 50 - Q07 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q07 a b 6 12 18 24 30 SE +/- 0.23, N = 3 24.86 25.86
Apache Spark TPC-H Scale Factor: 50 - Q06 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q06 b a 1.3282 2.6564 3.9846 5.3128 6.641 SE +/- 0.04522799, N = 3 5.88382483 5.90309207
Apache Spark TPC-H Scale Factor: 50 - Q05 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q05 a b 7 14 21 28 35 SE +/- 0.67, N = 3 29.84 31.20
Apache Spark TPC-H Scale Factor: 50 - Q04 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q04 a b 5 10 15 20 25 SE +/- 0.46, N = 3 21.00 21.82
Apache Spark TPC-H Scale Factor: 50 - Q03 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q03 a b 7 14 21 28 35 SE +/- 0.94, N = 3 26.19 29.69
Apache Spark TPC-H Scale Factor: 50 - Q02 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q02 a b 4 8 12 16 20 SE +/- 0.35, N = 3 14.25 14.53
Apache Spark TPC-H Scale Factor: 50 - Q01 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 50 - Q01 a b 3 6 9 12 15 SE +/- 0.21, N = 3 12.01 12.87
Apache Spark TPC-H Scale Factor: 10 - Q22 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q22 b a 2 4 6 8 10 SE +/- 0.13164085, N = 3 6.04430914 6.05411895
Apache Spark TPC-H Scale Factor: 10 - Q21 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q21 b a 8 16 24 32 40 SE +/- 0.11, N = 3 32.70 32.91
Apache Spark TPC-H Scale Factor: 10 - Q20 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q20 a b 3 6 9 12 15 SE +/- 0.14, N = 3 11.44 11.54
Apache Spark TPC-H Scale Factor: 10 - Q19 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q19 b a 2 4 6 8 10 SE +/- 0.15363169, N = 3 6.06041670 6.20677837
Apache Spark TPC-H Scale Factor: 10 - Q18 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q18 b a 5 10 15 20 25 SE +/- 0.50, N = 3 17.31 18.47
Apache Spark TPC-H Scale Factor: 10 - Q17 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q17 a b 3 6 9 12 15 SE +/- 0.07, N = 3 12.77 13.01
Apache Spark TPC-H Scale Factor: 10 - Q16 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q16 a b 2 4 6 8 10 SE +/- 0.33462632, N = 3 6.87131294 6.95270681
Apache Spark TPC-H Scale Factor: 10 - Q15 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q15 b a 1.3143 2.6286 3.9429 5.2572 6.5715 SE +/- 0.10221447, N = 3 5.43870592 5.84138076
Apache Spark TPC-H Scale Factor: 10 - Q14 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q14 b a 2 4 6 8 10 SE +/- 0.33271668, N = 3 6.90602303 7.07622369
Apache Spark TPC-H Scale Factor: 10 - Q13 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q13 a b 2 4 6 8 10 SE +/- 0.09689769, N = 3 7.37728373 7.94083786
Apache Spark TPC-H Scale Factor: 10 - Q12 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q12 a b 3 6 9 12 15 SE +/- 0.16460967, N = 3 9.94400438 10.03829002
Apache Spark TPC-H Scale Factor: 10 - Q11 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q11 a b 2 4 6 8 10 SE +/- 0.04584382, N = 3 8.00292349 8.27814293
Apache Spark TPC-H Scale Factor: 10 - Q10 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q10 b a 4 8 12 16 20 SE +/- 0.31, N = 3 14.78 15.17
Apache Spark TPC-H Scale Factor: 10 - Q09 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q09 a b 5 10 15 20 25 SE +/- 0.51, N = 3 21.91 22.53
Apache Spark TPC-H Scale Factor: 10 - Q08 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q08 b a 4 8 12 16 20 SE +/- 0.41, N = 3 14.56 15.52
Apache Spark TPC-H Scale Factor: 10 - Q07 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q07 a b 4 8 12 16 20 SE +/- 0.33, N = 3 14.65 14.90
Apache Spark TPC-H Scale Factor: 10 - Q06 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q06 b a 0.4615 0.923 1.3845 1.846 2.3075 SE +/- 0.23574646, N = 3 1.85595930 2.05104745
Apache Spark TPC-H Scale Factor: 10 - Q05 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q05 a b 5 10 15 20 25 SE +/- 0.48, N = 3 16.44 18.86
Apache Spark TPC-H Scale Factor: 10 - Q04 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q04 b a 3 6 9 12 15 SE +/- 0.21, N = 3 11.26 12.35
Apache Spark TPC-H Scale Factor: 10 - Q03 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q03 a b 4 8 12 16 20 SE +/- 0.31, N = 3 13.97 14.29
Apache Spark TPC-H Scale Factor: 10 - Q02 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q02 b a 2 4 6 8 10 SE +/- 0.13824959, N = 3 7.39245987 7.43104283
Apache Spark TPC-H Scale Factor: 10 - Q01 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 10 - Q01 b a 2 4 6 8 10 SE +/- 0.23898111, N = 3 7.28826714 7.58889151
Apache Spark TPC-H Scale Factor: 1 - Q22 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q22 a b 0.24 0.48 0.72 0.96 1.2 SE +/- 0.03222070, N = 3 1.00769047 1.06679213
Apache Spark TPC-H Scale Factor: 1 - Q21 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q21 b a 3 6 9 12 15 SE +/- 0.26119238, N = 3 9.55909538 9.64531231
Apache Spark TPC-H Scale Factor: 1 - Q20 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q20 b a 0.6879 1.3758 2.0637 2.7516 3.4395 SE +/- 0.12035470, N = 3 3.05001688 3.05739617
Apache Spark TPC-H Scale Factor: 1 - Q19 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q19 a b 0.193 0.386 0.579 0.772 0.965 SE +/- 0.03922839, N = 3 0.79092395 0.85797596
Apache Spark TPC-H Scale Factor: 1 - Q18 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q18 b a 1.2664 2.5328 3.7992 5.0656 6.332 SE +/- 0.11078356, N = 3 5.13171148 5.62853845
Apache Spark TPC-H Scale Factor: 1 - Q17 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q17 b a 0.666 1.332 1.998 2.664 3.33 SE +/- 0.10612827, N = 3 2.88348198 2.95993924
Apache Spark TPC-H Scale Factor: 1 - Q16 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q16 a b 0.3415 0.683 1.0245 1.366 1.7075 SE +/- 0.06760680, N = 3 1.38147259 1.51779914
Apache Spark TPC-H Scale Factor: 1 - Q15 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q15 a b 0.5821 1.1642 1.7463 2.3284 2.9105 SE +/- 0.11136502, N = 3 2.50185966 2.58714175
Apache Spark TPC-H Scale Factor: 1 - Q14 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q14 a b 0.4976 0.9952 1.4928 1.9904 2.488 SE +/- 0.16557850, N = 3 2.06485331 2.21146965
Apache Spark TPC-H Scale Factor: 1 - Q13 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q13 a b 0.3917 0.7834 1.1751 1.5668 1.9585 SE +/- 0.15789062, N = 3 1.58815936 1.74074161
Apache Spark TPC-H Scale Factor: 1 - Q12 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q12 a b 0.5099 1.0198 1.5297 2.0396 2.5495 SE +/- 0.15180813, N = 3 2.17542648 2.26641607
Apache Spark TPC-H Scale Factor: 1 - Q11 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q11 b a 0.2865 0.573 0.8595 1.146 1.4325 SE +/- 0.06007206, N = 3 1.13998687 1.27338135
Apache Spark TPC-H Scale Factor: 1 - Q10 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q10 b a 0.8581 1.7162 2.5743 3.4324 4.2905 SE +/- 0.13264795, N = 3 3.81245542 3.81359665
Apache Spark TPC-H Scale Factor: 1 - Q09 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q09 a b 1.327 2.654 3.981 5.308 6.635 SE +/- 0.08828966, N = 3 5.70969407 5.89775848
Apache Spark TPC-H Scale Factor: 1 - Q08 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q08 b a 0.5976 1.1952 1.7928 2.3904 2.988 SE +/- 0.02941830, N = 3 2.60907817 2.65584644
Apache Spark TPC-H Scale Factor: 1 - Q07 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q07 b a 0.9024 1.8048 2.7072 3.6096 4.512 SE +/- 0.02085439, N = 3 3.87790275 4.01044806
Apache Spark TPC-H Scale Factor: 1 - Q06 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q06 b a 0.1054 0.2108 0.3162 0.4216 0.527 SE +/- 0.03244463, N = 3 0.35801557 0.46822915
Apache Spark TPC-H Scale Factor: 1 - Q05 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q05 b a 0.9295 1.859 2.7885 3.718 4.6475 SE +/- 0.18898243, N = 3 3.69217634 4.13122161
Apache Spark TPC-H Scale Factor: 1 - Q04 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q04 b a 0.8832 1.7664 2.6496 3.5328 4.416 SE +/- 0.09899955, N = 3 3.75427246 3.92525745
Apache Spark TPC-H Scale Factor: 1 - Q03 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q03 a b 0.8699 1.7398 2.6097 3.4796 4.3495 SE +/- 0.11371323, N = 3 3.86442184 3.86610818
Apache Spark TPC-H Scale Factor: 1 - Q02 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q02 a b 0.4685 0.937 1.4055 1.874 2.3425 SE +/- 0.02016184, N = 3 2.06179071 2.08224201
Apache Spark TPC-H Scale Factor: 1 - Q01 OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark TPC-H 3.5 Scale Factor: 1 - Q01 a b 1.0005 2.001 3.0015 4.002 5.0025 SE +/- 0.17358727, N = 3 4.32006081 4.44657946
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.30 Backend: Eigen b a 150 300 450 600 750 SE +/- 17.59, N = 8 715 704 1. (CXX) g++ options: -flto -pthread
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.30 Backend: BLAS b a 200 400 600 800 1000 SE +/- 18.54, N = 9 871 853 1. (CXX) g++ options: -flto -pthread
Phoronix Test Suite v10.8.5