oktoberfest Tests for a future article. Intel Core i5-12600K testing with a ASUS PRIME Z690-P WIFI D4 (0605 BIOS) and ASUS Intel ADL-S GT1 15GB on Ubuntu 22.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2310296-PTS-OKTOBERF32&grs&sor .
oktoberfest Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL OpenCL Vulkan Compiler File-System Screen Resolution a b Intel Core i5-12600K @ 6.30GHz (10 Cores / 16 Threads) ASUS PRIME Z690-P WIFI D4 (0605 BIOS) Intel Device 7aa7 16GB 1000GB Western Digital WDS100T1X0E-00AFY0 ASUS Intel ADL-S GT1 15GB (1450MHz) Realtek ALC897 ASUS MG28U Realtek RTL8125 2.5GbE + Intel Device 7af0 Ubuntu 22.04 5.19.0-051900rc6daily20220716-generic (x86_64) GNOME Shell 42.1 X Server 1.21.1.3 + Wayland 4.6 Mesa 22.0.1 OpenCL 3.0 1.2.204 GCC 11.4.0 ext4 3840x2160 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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/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 Disk Details - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 Processor Details - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x2c - Thermald 2.4.9 Java Details - OpenJDK Runtime Environment (build 11.0.20.1+1-post-Ubuntu-0ubuntu122.04) Python Details - Python 3.10.12 Security 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected
oktoberfest onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU aom-av1: Speed 9 Realtime - Bosphorus 4K aom-av1: Speed 6 Realtime - Bosphorus 1080p aom-av1: Speed 8 Realtime - Bosphorus 1080p aom-av1: Speed 6 Realtime - Bosphorus 4K ncnn: CPU-v2-v2 - mobilenet-v2 aom-av1: Speed 9 Realtime - Bosphorus 1080p aom-av1: Speed 10 Realtime - Bosphorus 4K liquid-dsp: 16 - 256 - 512 ncnn: CPU - yolov4-tiny ncnn: CPU - vision_transformer aom-av1: Speed 10 Realtime - Bosphorus 1080p openradioss: Bumper Beam ncnn: CPU - alexnet aom-av1: Speed 11 Realtime - Bosphorus 4K svt-av1: Preset 13 - Bosphorus 1080p deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream memcached: 1:10 ncnn: CPU - FastestDet ncnn: CPU - resnet50 ncnn: CPU - mobilenet liquid-dsp: 16 - 256 - 57 openradioss: Cell Phone Drop Test deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream avifenc: 2 ncnn: CPU - mnasnet apache: 100 svt-av1: Preset 12 - Bosphorus 1080p liquid-dsp: 1 - 256 - 57 stress-ng: Poll ncnn: CPU - squeezenet_ssd ncnn: CPU - vgg16 aom-av1: Speed 11 Realtime - Bosphorus 1080p avifenc: 10, Lossless deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-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: ResNet-50, Sparse INT8 - Synchronous Single-Stream qmcpack: H4_ae deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream memcached: 1:100 openradioss: Bird Strike on Windshield ncnn: CPU - shufflenet-v2 ncnn: CPU - blazeface ncnn: CPU-v3-v3 - mobilenet-v3 stress-ng: Matrix 3D Math ncnn: CPU - regnety_400m dav1d: Chimera 1080p apache: 500 deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream qmcpack: simple-H2O liquid-dsp: 16 - 256 - 32 stress-ng: Mixed Scheduler onednn: IP Shapes 3D - f32 - CPU qmcpack: FeCO6_b3lyp_gms avifenc: 6 deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream quantlib: Multi-Threaded vvenc: Bosphorus 1080p - Fast deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream ncnn: CPU - efficientnet-b0 whisper-cpp: ggml-base.en - 2016 State of the Union stress-ng: Floating Point liquid-dsp: 4 - 256 - 57 tensorflow: CPU - 64 - ResNet-50 aom-av1: Speed 8 Realtime - Bosphorus 4K svt-av1: Preset 8 - Bosphorus 1080p liquid-dsp: 2 - 256 - 57 liquid-dsp: 8 - 256 - 512 embree: Pathtracer - Crown liquid-dsp: 8 - 256 - 32 apache: 1000 dav1d: Summer Nature 1080p liquid-dsp: 8 - 256 - 57 svt-av1: Preset 12 - Bosphorus 4K vvenc: Bosphorus 1080p - Faster svt-av1: Preset 4 - Bosphorus 4K stress-ng: Cloning deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream avifenc: 0 deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream liquid-dsp: 4 - 256 - 512 libxsmm: 64 stress-ng: AVL Tree deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream ncnn: CPU - googlenet whisper-cpp: ggml-medium.en - 2016 State of the Union deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream nginx: 100 tensorflow: CPU - 16 - ResNet-50 deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream apache: 200 ospray: gravity_spheres_volume/dim_512/pathtracer/real_time stress-ng: Hash vvenc: Bosphorus 4K - Faster deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream ospray-studio: 2 - 4K - 16 - Path Tracer - CPU embree: Pathtracer ISPC - Asian Dragon Obj openradioss: INIVOL and Fluid Structure Interaction Drop Container embree: Pathtracer - Asian Dragon nginx: 500 sqlite: 1 ospray: gravity_spheres_volume/dim_512/scivis/real_time vvenc: Bosphorus 4K - Fast embree: Pathtracer ISPC - Crown svt-av1: Preset 13 - Bosphorus 4K ncnn: CPU - resnet18 stress-ng: Vector Floating Point nginx: 1000 espeak: Text-To-Speech Synthesis cassandra: Writes deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream qmcpack: LiH_ae_MSD deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream ospray-studio: 1 - 1080p - 16 - Path Tracer - CPU deepsparse: ResNet-50, Baseline - Synchronous Single-Stream deepsparse: ResNet-50, Baseline - Synchronous Single-Stream blender: BMW27 - CPU-Only avifenc: 6, Lossless ospray: particle_volume/scivis/real_time openradioss: Rubber O-Ring Seal Installation deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream quantlib: Single-Threaded ospray: particle_volume/ao/real_time deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream sqlite: 2 palabos: 400 aom-av1: Speed 6 Two-Pass - Bosphorus 1080p deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream libxsmm: 128 qmcpack: O_ae_pyscf_UHF onednn: Deconvolution Batch shapes_1d - f32 - CPU z3: 2.smt2 embree: Pathtracer ISPC - Asian Dragon ospray-studio: 1 - 4K - 1 - Path Tracer - CPU dav1d: Summer Nature 4K hpcg: 104 104 104 - 60 deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream embree: Pathtracer - Asian Dragon Obj encode-opus: WAV To Opus Encode whisper-cpp: ggml-small.en - 2016 State of the Union ospray-studio: 2 - 1080p - 16 - Path Tracer - CPU easywave: e2Asean Grid + BengkuluSept2007 Source - 240 svt-av1: Preset 4 - Bosphorus 1080p deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream stress-ng: Pipe palabos: 100 ospray-studio: 3 - 1080p - 32 - Path Tracer - CPU ospray-studio: 1 - 1080p - 32 - Path Tracer - CPU ospray-studio: 3 - 1080p - 1 - Path Tracer - CPU deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream deepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Stream ospray-studio: 3 - 4K - 16 - Path Tracer - CPU ospray-studio: 2 - 1080p - 1 - Path Tracer - CPU deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream liquid-dsp: 2 - 256 - 512 dav1d: Chimera 1080p 10-bit ospray-studio: 3 - 4K - 1 - Path Tracer - CPU stress-ng: Vector Shuffle ospray: particle_volume/pathtracer/real_time libxsmm: 32 ospray: gravity_spheres_volume/dim_512/ao/real_time ospray-studio: 2 - 4K - 1 - Path Tracer - CPU onednn: Convolution Batch Shapes Auto - f32 - CPU svt-av1: Preset 8 - Bosphorus 4K liquid-dsp: 4 - 256 - 32 deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream aom-av1: Speed 6 Two-Pass - Bosphorus 4K ospray-studio: 1 - 4K - 16 - Path Tracer - CPU nginx: 200 deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream easywave: e2Asean Grid + BengkuluSept2007 Source - 1200 deepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream ospray-studio: 3 - 4K - 32 - Path Tracer - CPU tensorflow: CPU - 32 - ResNet-50 deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream blender: Pabellon Barcelona - CPU-Only ospray-studio: 3 - 1080p - 16 - Path Tracer - CPU deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream ospray-studio: 2 - 1080p - 32 - Path Tracer - CPU brl-cad: VGR Performance Metric stress-ng: Zlib deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Stream stress-ng: Fused Multiply-Add deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream ospray-studio: 1 - 1080p - 1 - Path Tracer - CPU deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream blender: Classroom - CPU-Only deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream nekrs: Kershaw deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream qmcpack: Li2_STO_ae blender: Fishy Cat - CPU-Only onednn: Deconvolution Batch shapes_3d - f32 - CPU deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream z3: 1.smt2 ospray-studio: 1 - 4K - 32 - Path Tracer - CPU onednn: IP Shapes 1D - f32 - CPU stress-ng: AVX-512 VNNI liquid-dsp: 1 - 256 - 512 stress-ng: Wide Vector Math build-gcc: Time To Compile stress-ng: Pthread liquid-dsp: 2 - 256 - 32 blender: Barbershop - CPU-Only sqlite: 4 ospray-studio: 2 - 4K - 32 - Path Tracer - CPU liquid-dsp: 1 - 256 - 32 duckdb: TPC-H Parquet duckdb: IMDB build2: Time To Compile build-godot: Time To Compile openvkl: vklBenchmarkCPU Scalar openvkl: vklBenchmarkCPU ISPC oidn: RT.ldr_alb_nrm.3840x2160 - CPU-Only aom-av1: Speed 4 Two-Pass - Bosphorus 1080p aom-av1: Speed 0 Two-Pass - Bosphorus 1080p aom-av1: Speed 4 Two-Pass - Bosphorus 4K aom-av1: Speed 0 Two-Pass - Bosphorus 4K nekrs: TurboPipe Periodic hpcg: 144 144 144 - 60 a b 5530.13 2636.78 69.71 146.21 183.46 63.82 2.98 199.96 75.67 185450000 16.64 100.28 178.89 222.32 4.96 73.19 516.2 45.4754 21.986 3279849.87 3 12.77 10.24 695800000 157.04 19.6842 50.7825 60.818 2.54 148966.44 437.372 78801000 1241130.3 6.86 37.31 193.68 4.552 2.8555 87.647 11.4028 349.6151 39.88 472.6798 3016454.9 347.26 2.24 0.75 2.38 1401.51 5.98 683.61 119690.62 10.4405 8.4541 9.7413 102.6438 43.891 757790000 10186.26 10.7972 341.81 6.332 49.7154 69.7199 588.3009 55.415 18.0424 40312.3 15.108 71.6364 100.4053 4.87 244.72794 4377.45 305790000 17.81 68.67 92.681 159380000 130420000 12.706 402980000 120445.38 956.18 518790000 109.879 33.291 3.475 860.48 35.367 28.2718 127.473 164.2155 80680000 167.9 81.66 30.4378 7.77 2723.0535 17.7976 56.1658 108258.46 17.01 7.754 128.9608 140860.12 3.88095 2294409.16 10.181 586.6719 162070 14.513 717.69 14.9731 89266.33 8.943 2.55887 4.931 13.4891 111.129 5.52 19538.35 82318.98 21.215 123080 52.3238 201.74 95.4789 42656 79.1508 12.6294 112.43 9.164 5.68706 283.82 22.4104 4562.6 5.74695 9.656 14.49 76.7579 46.78 222.916 98.0783 50.9696 114.5198 247.5 314.45 9.24791 62.939 16.6603 9734 238.78 6.95045 144.1229 13.5052 20.508 820.19569 43199 9.046 11.198 7.0052 34.6065 7171062.72 51.2381 96823 82003 2912 712.3593 22.0647 45.3071 189279 2499 7.7535 128.9697 41548000 556.38 11549 9447.65 140.441 96.3 2.63693 9910 14.3545 44.977 221860000 109.1421 14.25 159663 102547.95 12.6553 185.879 78.9909 375733 17.49 43.6378 385.7 49884 46.1862 108.1187 83367 215292 1130.39 8.145 122.7701 16384409.66 511.3289 2456 45.7934 316.24 21.1919 3261960000 47.1806 445.38 156.02 8.04899 8.4527 19.126 316965 3.89503 1228899.23 20784000 484455.75 918.832 186367.1 115700000 1237.27 19.005 322602 57879000 97.820 115.408 137.603 323.861 130 277 0.34 15.99 0.73 6.77 0.24 4270.3 2167.5 79.9 166.68 162.27 71.58 3.34 184.04 81.38 196230000 17.59 95.33 188.01 233.21 5.14 70.73 499.07 44.0455 22.6997 3183290.59 3.09 13.15 10.54 676010000 152.73 20.2367 49.3977 59.356 2.6 145590.45 427.53 80378000 1217218.91 6.99 36.62 190.21 4.634 2.8079 89.1195 11.2144 355.4745 40.51 479.6106 3059379.28 342.5 2.27 0.76 2.41 1418.21 6.05 675.83 121041.48 10.3248 8.3633 9.6372 103.7525 44.358 750070000 10290.81 10.6941 338.6 6.274 50.1726 69.0868 593.639 54.9374 18.1991 39967.5 15.237 72.2413 99.5737 4.91 246.69041 4412.33 303380000 17.95 68.16 93.344 160510000 131280000 12.623 400370000 119665.35 962.35 515540000 109.209 33.094 3.455 865.39 35.5633 28.1161 126.773 163.3215 81117000 167 82.1 30.5986 7.73 2736.77075 17.7105 56.4406 107751.55 17.09 7.7181 129.5604 140213.08 3.86334 2304545.32 10.225 589.1531 162754 14.5736 714.81 14.9131 88913.24 8.908 2.54893 4.912 13.5405 111.541 5.5 19468.51 82032.62 21.288 122667 52.4969 202.34 95.2027 42536 79.3733 12.5944 112.12 9.139 5.70261 283.06 22.3515 4573.9 5.76091 9.679 14.457 76.5863 46.68 223.3921 98.2865 50.8618 114.2856 248 313.84 9.2303 63.059 16.6913 9716 239.22 6.96238 144.3669 13.4833 20.541 818.91356 43132 9.06 11.215 6.9947 34.5555 7160556.83 51.3091 96957 82116 2916 713.3158 22.0939 45.2474 189045 2496 7.7442 129.1244 41499000 555.73 11536 9457.69 140.588 96.4 2.63451 9901 14.3415 44.94 221680000 109.0587 14.26 159770 102479.68 12.6469 185.999 79.0416 375965 17.5 43.6626 385.91 49910 46.2101 108.0654 83406 215193 1129.89 8.1485 122.7176 16391410.53 511.1198 2457 45.8118 316.35 21.1846 3260840000 47.1966 445.23 156.07 8.04648 8.4551 19.121 317047 3.89409 1229104.91 20781000 484394.85 918.938 186386.61 115690000 1237.36 19.004 322586 57881000 130 277 0.34 15.99 0.73 6.77 0.24 4079830000 OpenBenchmarking.org
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU b a 1200 2400 3600 4800 6000 4270.30 5530.13 MIN: 4261.15 MIN: 4272.6 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU b a 600 1200 1800 2400 3000 2167.50 2636.78 MIN: 2155.94 MIN: 2149.96 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K b a 20 40 60 80 100 79.90 69.71 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p b a 40 80 120 160 200 166.68 146.21 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p a b 40 80 120 160 200 183.46 162.27 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K b a 16 32 48 64 80 71.58 63.82 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
NCNN Target: CPU-v2-v2 - Model: mobilenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU-v2-v2 - Model: mobilenet-v2 a b 0.7515 1.503 2.2545 3.006 3.7575 2.98 3.34 MIN: 2.96 / MAX: 3.1 MIN: 3.33 / MAX: 3.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p a b 40 80 120 160 200 199.96 184.04 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K b a 20 40 60 80 100 81.38 75.67 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Liquid-DSP Threads: 16 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 512 b a 40M 80M 120M 160M 200M 196230000 185450000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
NCNN Target: CPU - Model: yolov4-tiny OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: yolov4-tiny a b 4 8 12 16 20 16.64 17.59 MIN: 16.5 / MAX: 16.86 MIN: 17.43 / MAX: 23.14 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vision_transformer OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: vision_transformer b a 20 40 60 80 100 95.33 100.28 MIN: 94.07 / MAX: 152.19 MIN: 94 / MAX: 163.09 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p b a 40 80 120 160 200 188.01 178.89 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenRadioss Model: Bumper Beam OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bumper Beam a b 50 100 150 200 250 222.32 233.21
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: alexnet a b 1.1565 2.313 3.4695 4.626 5.7825 4.96 5.14 MIN: 4.89 / MAX: 5.09 MIN: 5.08 / MAX: 5.22 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 4K a b 16 32 48 64 80 73.19 70.73 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 1080p a b 110 220 330 440 550 516.20 499.07 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream a b 10 20 30 40 50 45.48 44.05
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 21.99 22.70
Memcached Set To Get Ratio: 1:10 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:10 a b 700K 1400K 2100K 2800K 3500K 3279849.87 3183290.59 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
NCNN Target: CPU - Model: FastestDet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: FastestDet a b 0.6953 1.3906 2.0859 2.7812 3.4765 3.00 3.09 MIN: 2.96 / MAX: 3.14 MIN: 3.08 / MAX: 3.18 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: resnet50 a b 3 6 9 12 15 12.77 13.15 MIN: 12.66 / MAX: 13.17 MIN: 12.98 / MAX: 13.35 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: mobilenet a b 3 6 9 12 15 10.24 10.54 MIN: 10.16 / MAX: 10.54 MIN: 10.47 / MAX: 10.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Liquid-DSP Threads: 16 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 57 a b 150M 300M 450M 600M 750M 695800000 676010000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenRadioss Model: Cell Phone Drop Test OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Cell Phone Drop Test b a 30 60 90 120 150 152.73 157.04
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 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 19.68 20.24
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 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream a b 11 22 33 44 55 50.78 49.40
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 2 b a 14 28 42 56 70 59.36 60.82 1. (CXX) g++ options: -O3 -fPIC -lm
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: mnasnet a b 0.585 1.17 1.755 2.34 2.925 2.54 2.60 MIN: 2.52 / MAX: 2.6 MIN: 2.58 / MAX: 2.73 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Apache HTTP Server Concurrent Requests: 100 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 100 a b 30K 60K 90K 120K 150K 148966.44 145590.45 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 1080p a b 90 180 270 360 450 437.37 427.53 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 57 b a 20M 40M 60M 80M 100M 80378000 78801000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Test: Poll OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Poll a b 300K 600K 900K 1200K 1500K 1241130.30 1217218.91 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
NCNN Target: CPU - Model: squeezenet_ssd OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: squeezenet_ssd a b 2 4 6 8 10 6.86 6.99 MIN: 6.78 / MAX: 6.97 MIN: 6.91 / MAX: 7.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vgg16 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: vgg16 b a 9 18 27 36 45 36.62 37.31 MIN: 36.48 / MAX: 43.16 MIN: 37.22 / MAX: 37.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
AOM AV1 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 11 Realtime - Input: Bosphorus 1080p a b 40 80 120 160 200 193.68 190.21 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 10, Lossless a b 1.0427 2.0854 3.1281 4.1708 5.2135 4.552 4.634 1. (CXX) g++ options: -O3 -fPIC -lm
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream b a 0.6425 1.285 1.9275 2.57 3.2125 2.8079 2.8555
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 1.5 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream b a 20 40 60 80 100 89.12 87.65
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 1.5 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream b a 3 6 9 12 15 11.21 11.40
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream b a 80 160 240 320 400 355.47 349.62
QMCPACK Input: H4_ae OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: H4_ae a b 9 18 27 36 45 39.88 40.51 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a b 100 200 300 400 500 472.68 479.61
Memcached Set To Get Ratio: 1:100 OpenBenchmarking.org Ops/sec, More Is Better Memcached 1.6.19 Set To Get Ratio: 1:100 b a 700K 1400K 2100K 2800K 3500K 3059379.28 3016454.90 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenRadioss Model: Bird Strike on Windshield OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Bird Strike on Windshield b a 80 160 240 320 400 342.50 347.26
NCNN Target: CPU - Model: shufflenet-v2 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: shufflenet-v2 a b 0.5108 1.0216 1.5324 2.0432 2.554 2.24 2.27 MIN: 2.23 / MAX: 2.28 MIN: 2.26 / MAX: 2.55 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: blazeface a b 0.171 0.342 0.513 0.684 0.855 0.75 0.76 MIN: 0.74 / MAX: 0.78 MIN: 0.75 / MAX: 0.77 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
NCNN Target: CPU-v3-v3 - Model: mobilenet-v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU-v3-v3 - Model: mobilenet-v3 a b 0.5423 1.0846 1.6269 2.1692 2.7115 2.38 2.41 MIN: 2.36 / MAX: 2.6 MIN: 2.38 / MAX: 2.55 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Stress-NG Test: Matrix 3D Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Matrix 3D Math b a 300 600 900 1200 1500 1418.21 1401.51 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
NCNN Target: CPU - Model: regnety_400m OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: regnety_400m a b 2 4 6 8 10 5.98 6.05 MIN: 5.94 / MAX: 6.11 MIN: 6.02 / MAX: 6.18 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 1.2.1 Video Input: Chimera 1080p a b 150 300 450 600 750 683.61 675.83 1. (CC) gcc options: -pthread -lm
Apache HTTP Server Concurrent Requests: 500 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 500 b a 30K 60K 90K 120K 150K 121041.48 119690.62 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream a b 3 6 9 12 15 10.44 10.32
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 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a b 2 4 6 8 10 8.4541 8.3633
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b 3 6 9 12 15 9.7413 9.6372
Neural Magic DeepSparse Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream a b 20 40 60 80 100 102.64 103.75
QMCPACK Input: simple-H2O OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: simple-H2O a b 10 20 30 40 50 43.89 44.36 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
Liquid-DSP Threads: 16 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 32 a b 160M 320M 480M 640M 800M 757790000 750070000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Test: Mixed Scheduler OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Mixed Scheduler b a 2K 4K 6K 8K 10K 10290.81 10186.26 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU b a 3 6 9 12 15 10.69 10.80 MIN: 10.6 MIN: 10.69 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
QMCPACK Input: FeCO6_b3lyp_gms OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: FeCO6_b3lyp_gms b a 70 140 210 280 350 338.60 341.81 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6 b a 2 4 6 8 10 6.274 6.332 1. (CXX) g++ options: -O3 -fPIC -lm
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream a b 11 22 33 44 55 49.72 50.17
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a b 16 32 48 64 80 69.72 69.09
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 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream a b 130 260 390 520 650 588.30 593.64
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a b 12 24 36 48 60 55.42 54.94
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream a b 4 8 12 16 20 18.04 18.20
QuantLib Configuration: Multi-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Multi-Threaded a b 9K 18K 27K 36K 45K 40312.3 39967.5 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
VVenC Video Input: Bosphorus 1080p - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Fast b a 4 8 12 16 20 15.24 15.11 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
Neural Magic DeepSparse Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream a b 16 32 48 64 80 71.64 72.24
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream a b 20 40 60 80 100 100.41 99.57
NCNN Target: CPU - Model: efficientnet-b0 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: efficientnet-b0 a b 1.1048 2.2096 3.3144 4.4192 5.524 4.87 4.91 MIN: 4.83 / MAX: 5 MIN: 4.86 / MAX: 5.12 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Whisper.cpp Model: ggml-base.en - Input: 2016 State of the Union OpenBenchmarking.org Seconds, Fewer Is Better Whisper.cpp 1.4 Model: ggml-base.en - Input: 2016 State of the Union a b 50 100 150 200 250 244.73 246.69 1. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread
Stress-NG Test: Floating Point OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Floating Point b a 900 1800 2700 3600 4500 4412.33 4377.45 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Liquid-DSP Threads: 4 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 57 a b 70M 140M 210M 280M 350M 305790000 303380000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 b a 4 8 12 16 20 17.95 17.81
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K a b 15 30 45 60 75 68.67 68.16 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 1080p b a 20 40 60 80 100 93.34 92.68 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Liquid-DSP Threads: 2 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 57 b a 30M 60M 90M 120M 150M 160510000 159380000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Liquid-DSP Threads: 8 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 512 b a 30M 60M 90M 120M 150M 131280000 130420000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Crown a b 3 6 9 12 15 12.71 12.62 MIN: 12.62 / MAX: 12.94 MIN: 12.52 / MAX: 12.78
Liquid-DSP Threads: 8 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 32 a b 90M 180M 270M 360M 450M 402980000 400370000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Apache HTTP Server Concurrent Requests: 1000 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 1000 a b 30K 60K 90K 120K 150K 120445.38 119665.35 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 1.2.1 Video Input: Summer Nature 1080p b a 200 400 600 800 1000 962.35 956.18 1. (CC) gcc options: -pthread -lm
Liquid-DSP Threads: 8 - Buffer Length: 256 - Filter Length: 57 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 57 a b 110M 220M 330M 440M 550M 518790000 515540000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 12 - Input: Bosphorus 4K a b 20 40 60 80 100 109.88 109.21 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
VVenC Video Input: Bosphorus 1080p - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 1080p - Video Preset: Faster a b 8 16 24 32 40 33.29 33.09 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K a b 0.7819 1.5638 2.3457 3.1276 3.9095 3.475 3.455 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Stress-NG Test: Cloning OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Cloning b a 200 400 600 800 1000 865.39 860.48 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a b 8 16 24 32 40 35.37 35.56
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream a b 7 14 21 28 35 28.27 28.12
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 0 b a 30 60 90 120 150 126.77 127.47 1. (CXX) g++ options: -O3 -fPIC -lm
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream b a 40 80 120 160 200 163.32 164.22
Liquid-DSP Threads: 4 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 512 b a 20M 40M 60M 80M 100M 81117000 80680000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
libxsmm M N K: 64 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 64 a b 40 80 120 160 200 167.9 167.0 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
Stress-NG Test: AVL Tree OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: AVL Tree b a 20 40 60 80 100 82.10 81.66 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream b a 7 14 21 28 35 30.60 30.44
NCNN Target: CPU - Model: googlenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: googlenet b a 2 4 6 8 10 7.73 7.77 MIN: 7.62 / MAX: 7.86 MIN: 7.62 / MAX: 7.91 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Whisper.cpp Model: ggml-medium.en - Input: 2016 State of the Union OpenBenchmarking.org Seconds, Fewer Is Better Whisper.cpp 1.4 Model: ggml-medium.en - Input: 2016 State of the Union a b 600 1200 1800 2400 3000 2723.05 2736.77 1. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream b a 4 8 12 16 20 17.71 17.80
Neural Magic DeepSparse Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream b a 13 26 39 52 65 56.44 56.17
nginx Connections: 100 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 100 a b 20K 40K 60K 80K 100K 108258.46 107751.55 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 b a 4 8 12 16 20 17.09 17.01
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 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b 2 4 6 8 10 7.7540 7.7181
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 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream a b 30 60 90 120 150 128.96 129.56
Apache HTTP Server Concurrent Requests: 200 OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.56 Concurrent Requests: 200 a b 30K 60K 90K 120K 150K 140860.12 140213.08 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OSPRay Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time a b 0.8732 1.7464 2.6196 3.4928 4.366 3.88095 3.86334
Stress-NG Test: Hash OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Hash b a 500K 1000K 1500K 2000K 2500K 2304545.32 2294409.16 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
VVenC Video Input: Bosphorus 4K - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Faster b a 3 6 9 12 15 10.23 10.18 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
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 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream a b 130 260 390 520 650 586.67 589.15
OSPRay Studio Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU a b 30K 60K 90K 120K 150K 162070 162754
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj b a 4 8 12 16 20 14.57 14.51 MIN: 14.49 / MAX: 14.75 MIN: 14.43 / MAX: 14.71
OpenRadioss Model: INIVOL and Fluid Structure Interaction Drop Container OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: INIVOL and Fluid Structure Interaction Drop Container b a 150 300 450 600 750 714.81 717.69
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Asian Dragon a b 4 8 12 16 20 14.97 14.91 MIN: 14.91 / MAX: 15.16 MIN: 14.84 / MAX: 15.18
nginx Connections: 500 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 a b 20K 40K 60K 80K 100K 89266.33 88913.24 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
SQLite Threads / Copies: 1 OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.41.2 Threads / Copies: 1 b a 2 4 6 8 10 8.908 8.943 1. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm
OSPRay Benchmark: gravity_spheres_volume/dim_512/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time a b 0.5757 1.1514 1.7271 2.3028 2.8785 2.55887 2.54893
VVenC Video Input: Bosphorus 4K - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.9 Video Input: Bosphorus 4K - Video Preset: Fast a b 1.1095 2.219 3.3285 4.438 5.5475 4.931 4.912 1. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Crown b a 3 6 9 12 15 13.54 13.49 MIN: 13.42 / MAX: 13.73 MIN: 13.36 / MAX: 13.73
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 13 - Input: Bosphorus 4K b a 20 40 60 80 100 111.54 111.13 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
NCNN Target: CPU - Model: resnet18 OpenBenchmarking.org ms, Fewer Is Better NCNN 20230517 Target: CPU - Model: resnet18 b a 1.242 2.484 3.726 4.968 6.21 5.50 5.52 MIN: 5.39 / MAX: 5.62 MIN: 5.35 / MAX: 5.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
Stress-NG Test: Vector Floating Point OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Floating Point a b 4K 8K 12K 16K 20K 19538.35 19468.51 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
nginx Connections: 1000 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 a b 20K 40K 60K 80K 100K 82318.98 82032.62 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
eSpeak-NG Speech Engine Text-To-Speech Synthesis OpenBenchmarking.org Seconds, Fewer Is Better eSpeak-NG Speech Engine 1.51 Text-To-Speech Synthesis a b 5 10 15 20 25 21.22 21.29 1. (CXX) g++ options: -O2
Apache Cassandra Test: Writes OpenBenchmarking.org Op/s, More Is Better Apache Cassandra 4.1.3 Test: Writes a b 30K 60K 90K 120K 150K 123080 122667
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream b a 12 24 36 48 60 52.50 52.32
QMCPACK Input: LiH_ae_MSD OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: LiH_ae_MSD a b 40 80 120 160 200 201.74 202.34 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream b a 20 40 60 80 100 95.20 95.48
OSPRay Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU b a 9K 18K 27K 36K 45K 42536 42656
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream b a 20 40 60 80 100 79.37 79.15
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream b a 3 6 9 12 15 12.59 12.63
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: BMW27 - Compute: CPU-Only b a 30 60 90 120 150 112.12 112.43
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 1.0 Encoder Speed: 6, Lossless b a 3 6 9 12 15 9.139 9.164 1. (CXX) g++ options: -O3 -fPIC -lm
OSPRay Benchmark: particle_volume/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/scivis/real_time b a 1.2831 2.5662 3.8493 5.1324 6.4155 5.70261 5.68706
OpenRadioss Model: Rubber O-Ring Seal Installation OpenBenchmarking.org Seconds, Fewer Is Better OpenRadioss 2023.09.15 Model: Rubber O-Ring Seal Installation b a 60 120 180 240 300 283.06 283.82
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 1.5 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 5 10 15 20 25 22.35 22.41
QuantLib Configuration: Single-Threaded OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.32 Configuration: Single-Threaded b a 1000 2000 3000 4000 5000 4573.9 4562.6 1. (CXX) g++ options: -O3 -march=native -fPIE -pie
OSPRay Benchmark: particle_volume/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/ao/real_time b a 1.2962 2.5924 3.8886 5.1848 6.481 5.76091 5.74695
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream b a 3 6 9 12 15 9.679 9.656
SQLite Threads / Copies: 2 OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.41.2 Threads / Copies: 2 b a 4 8 12 16 20 14.46 14.49 1. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm
Palabos Grid Size: 400 OpenBenchmarking.org Mega Site Updates Per Second, More Is Better Palabos 2.3 Grid Size: 400 a b 20 40 60 80 100 76.76 76.59 1. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p a b 11 22 33 44 55 46.78 46.68 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
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 1.5 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 50 100 150 200 250 223.39 222.92
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 20 40 60 80 100 98.08 98.29
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 11 22 33 44 55 50.97 50.86
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 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 30 60 90 120 150 114.52 114.29
libxsmm M N K: 128 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 128 b a 50 100 150 200 250 248.0 247.5 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
QMCPACK Input: O_ae_pyscf_UHF OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: O_ae_pyscf_UHF b a 70 140 210 280 350 313.84 314.45 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU b a 3 6 9 12 15 9.23030 9.24791 MIN: 5.49 MIN: 5.33 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Z3 Theorem Prover SMT File: 2.smt2 OpenBenchmarking.org Seconds, Fewer Is Better Z3 Theorem Prover 4.12.1 SMT File: 2.smt2 a b 14 28 42 56 70 62.94 63.06 1. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon b a 4 8 12 16 20 16.69 16.66 MIN: 16.6 / MAX: 16.96 MIN: 16.56 / MAX: 16.95
OSPRay Studio Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU b a 2K 4K 6K 8K 10K 9716 9734
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 1.2.1 Video Input: Summer Nature 4K b a 50 100 150 200 250 239.22 238.78 1. (CC) gcc options: -pthread -lm
High Performance Conjugate Gradient X Y Z: 104 104 104 - RT: 60 OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 b a 2 4 6 8 10 6.96238 6.95045 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a b 30 60 90 120 150 144.12 144.37
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.3 Binary: Pathtracer - Model: Asian Dragon Obj a b 3 6 9 12 15 13.51 13.48 MIN: 13.43 / MAX: 13.66 MIN: 13.41 / MAX: 13.7
Opus Codec Encoding WAV To Opus Encode OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.4 WAV To Opus Encode a b 5 10 15 20 25 20.51 20.54 1. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
Whisper.cpp Model: ggml-small.en - Input: 2016 State of the Union OpenBenchmarking.org Seconds, Fewer Is Better Whisper.cpp 1.4 Model: ggml-small.en - Input: 2016 State of the Union b a 200 400 600 800 1000 818.91 820.20 1. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread
OSPRay Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU b a 9K 18K 27K 36K 45K 43132 43199
easyWave Input: e2Asean Grid + BengkuluSept2007 Source - Time: 240 OpenBenchmarking.org Seconds, Fewer Is Better easyWave r34 Input: e2Asean Grid + BengkuluSept2007 Source - Time: 240 a b 3 6 9 12 15 9.046 9.060 1. (CXX) g++ options: -O3 -fopenmp
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 4 - Input: Bosphorus 1080p b a 3 6 9 12 15 11.22 11.20 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 2 4 6 8 10 6.9947 7.0052
Neural Magic DeepSparse Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream a b 8 16 24 32 40 34.61 34.56
Stress-NG Test: Pipe OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Pipe a b 1.5M 3M 4.5M 6M 7.5M 7171062.72 7160556.83 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Palabos Grid Size: 100 OpenBenchmarking.org Mega Site Updates Per Second, More Is Better Palabos 2.3 Grid Size: 100 b a 12 24 36 48 60 51.31 51.24 1. (CXX) g++ options: -std=c++17 -pedantic -O3 -rdynamic -lcrypto -lcurl -lsz -lz -ldl -lm
OSPRay Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU a b 20K 40K 60K 80K 100K 96823 96957
OSPRay Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU a b 20K 40K 60K 80K 100K 82003 82116
OSPRay Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU a b 600 1200 1800 2400 3000 2912 2916
Neural Magic DeepSparse Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream b a 150 300 450 600 750 713.32 712.36
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 22.06 22.09
Neural Magic DeepSparse Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream a b 10 20 30 40 50 45.31 45.25
OSPRay Studio Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU b a 40K 80K 120K 160K 200K 189045 189279
OSPRay Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU b a 500 1000 1500 2000 2500 2496 2499
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 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b 2 4 6 8 10 7.7535 7.7442
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 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream a b 30 60 90 120 150 128.97 129.12
Liquid-DSP Threads: 2 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 512 a b 9M 18M 27M 36M 45M 41548000 41499000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 1.2.1 Video Input: Chimera 1080p 10-bit a b 120 240 360 480 600 556.38 555.73 1. (CC) gcc options: -pthread -lm
OSPRay Studio Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU b a 2K 4K 6K 8K 10K 11536 11549
Stress-NG Test: Vector Shuffle OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Shuffle b a 2K 4K 6K 8K 10K 9457.69 9447.65 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
OSPRay Benchmark: particle_volume/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time b a 30 60 90 120 150 140.59 140.44
libxsmm M N K: 32 OpenBenchmarking.org GFLOPS/s, More Is Better libxsmm 2-1.17-3645 M N K: 32 b a 20 40 60 80 100 96.4 96.3 1. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2
OSPRay Benchmark: gravity_spheres_volume/dim_512/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time a b 0.5933 1.1866 1.7799 2.3732 2.9665 2.63693 2.63451
OSPRay Studio Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU b a 2K 4K 6K 8K 10K 9901 9910
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU b a 4 8 12 16 20 14.34 14.35 MIN: 14.17 MIN: 14.15 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K a b 10 20 30 40 50 44.98 44.94 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Liquid-DSP Threads: 4 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 32 a b 50M 100M 150M 200M 250M 221860000 221680000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream a b 20 40 60 80 100 109.14 109.06
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K b a 4 8 12 16 20 14.26 14.25 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OSPRay Studio Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU a b 30K 60K 90K 120K 150K 159663 159770
nginx Connections: 200 OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 200 a b 20K 40K 60K 80K 100K 102547.95 102479.68 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream b a 3 6 9 12 15 12.65 12.66
easyWave Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200 OpenBenchmarking.org Seconds, Fewer Is Better easyWave r34 Input: e2Asean Grid + BengkuluSept2007 Source - Time: 1200 a b 40 80 120 160 200 185.88 186.00 1. (CXX) g++ options: -O3 -fopenmp
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream b a 20 40 60 80 100 79.04 78.99
OSPRay Studio Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU a b 80K 160K 240K 320K 400K 375733 375965
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 b a 4 8 12 16 20 17.50 17.49
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 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream a b 10 20 30 40 50 43.64 43.66
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Pabellon Barcelona - Compute: CPU-Only a b 80 160 240 320 400 385.70 385.91
OSPRay Studio Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU a b 11K 22K 33K 44K 55K 49884 49910
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b 10 20 30 40 50 46.19 46.21
Neural Magic DeepSparse Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream a b 20 40 60 80 100 108.12 108.07
OSPRay Studio Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU a b 20K 40K 60K 80K 100K 83367 83406
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.36 VGR Performance Metric a b 50K 100K 150K 200K 250K 215292 215193 1. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6
Stress-NG Test: Zlib OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Zlib a b 200 400 600 800 1000 1130.39 1129.89 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream b a 2 4 6 8 10 8.1485 8.1450
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Stream b a 30 60 90 120 150 122.72 122.77
Stress-NG Test: Fused Multiply-Add OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Fused Multiply-Add b a 4M 8M 12M 16M 20M 16391410.53 16384409.66 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Neural Magic DeepSparse Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream b a 110 220 330 440 550 511.12 511.33
OSPRay Studio Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU a b 500 1000 1500 2000 2500 2456 2457
Neural Magic DeepSparse Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream a b 10 20 30 40 50 45.79 45.81
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Classroom - Compute: CPU-Only a b 70 140 210 280 350 316.24 316.35
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a b 5 10 15 20 25 21.19 21.18
nekRS Input: Kershaw OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: Kershaw a b 700M 1400M 2100M 2800M 3500M 3261960000 3260840000 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi
Neural Magic DeepSparse Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream a b 11 22 33 44 55 47.18 47.20
QMCPACK Input: Li2_STO_ae OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.17.1 Input: Li2_STO_ae b a 100 200 300 400 500 445.23 445.38 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -march=native -O3 -lm -ldl
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Fishy Cat - Compute: CPU-Only a b 30 60 90 120 150 156.02 156.07
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU b a 2 4 6 8 10 8.04648 8.04899 MIN: 8.02 MIN: 8.02 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
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 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream b a 2 4 6 8 10 8.4551 8.4527
Z3 Theorem Prover SMT File: 1.smt2 OpenBenchmarking.org Seconds, Fewer Is Better Z3 Theorem Prover 4.12.1 SMT File: 1.smt2 b a 5 10 15 20 25 19.12 19.13 1. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC
OSPRay Studio Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU a b 70K 140K 210K 280K 350K 316965 317047
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 3.3 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU b a 0.8764 1.7528 2.6292 3.5056 4.382 3.89409 3.89503 MIN: 3.75 MIN: 3.75 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
Stress-NG Test: AVX-512 VNNI OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: AVX-512 VNNI b a 300K 600K 900K 1200K 1500K 1229104.91 1228899.23 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 512 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 a b 4M 8M 12M 16M 20M 20784000 20781000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Test: Wide Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Wide Vector Math a b 100K 200K 300K 400K 500K 484455.75 484394.85 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 13.2 Time To Compile a b 200 400 600 800 1000 918.83 918.94
Stress-NG Test: Pthread OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Pthread b a 40K 80K 120K 160K 200K 186386.61 186367.10 1. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lsctp -lz
Liquid-DSP Threads: 2 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 32 a b 20M 40M 60M 80M 100M 115700000 115690000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Barbershop - Compute: CPU-Only a b 300 600 900 1200 1500 1237.27 1237.36
SQLite Threads / Copies: 4 OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.41.2 Threads / Copies: 4 b a 5 10 15 20 25 19.00 19.01 1. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm
OSPRay Studio Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.13 Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU b a 70K 140K 210K 280K 350K 322586 322602
Liquid-DSP Threads: 1 - Buffer Length: 256 - Filter Length: 32 OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 b a 12M 24M 36M 48M 60M 57881000 57879000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
DuckDB Benchmark: TPC-H Parquet OpenBenchmarking.org Seconds, Fewer Is Better DuckDB 0.9.1 Benchmark: TPC-H Parquet a 20 40 60 80 100 SE +/- 0.18, N = 3 97.82 1. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl
DuckDB Benchmark: IMDB OpenBenchmarking.org Seconds, Fewer Is Better DuckDB 0.9.1 Benchmark: IMDB a 30 60 90 120 150 SE +/- 0.15, N = 3 115.41 1. (CXX) g++ options: -O3 -rdynamic -lssl -lcrypto -ldl
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.15 Time To Compile a 30 60 90 120 150 137.60
Timed Godot Game Engine Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 4.0 Time To Compile a 70 140 210 280 350 323.86
OpenVKL Benchmark: vklBenchmarkCPU Scalar OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU Scalar b a 30 60 90 120 150 130 130 MIN: 10 / MAX: 2218 MIN: 10 / MAX: 2222
OpenVKL Benchmark: vklBenchmarkCPU ISPC OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU ISPC b a 60 120 180 240 300 277 277 MIN: 20 / MAX: 3808 MIN: 20 / MAX: 3813
Intel Open Image Denoise Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.1 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only b a 0.0765 0.153 0.2295 0.306 0.3825 0.34 0.34
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p b a 4 8 12 16 20 15.99 15.99 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p b a 0.1643 0.3286 0.4929 0.6572 0.8215 0.73 0.73 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K b a 2 4 6 8 10 6.77 6.77 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.7 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K b a 0.054 0.108 0.162 0.216 0.27 0.24 0.24 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
nekRS Input: TurboPipe Periodic OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: TurboPipe Periodic b 900M 1800M 2700M 3600M 4500M 4079830000 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi
Phoronix Test Suite v10.8.5