9600X Benchmarks for a future article. AMD Ryzen 5 9600X 6-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (2308 BIOS) and AMD Radeon RX 7900 GRE 16GB on Ubuntu 24.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2409118-PTS-9600X70297&grs&sro .
9600X Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution a b AMD Ryzen 5 9600X 6-Core @ 5.48GHz (6 Cores / 12 Threads) ASUS ROG STRIX X670E-E GAMING WIFI (2308 BIOS) AMD Device 14d8 2 x 16GB DDR5-6000MT/s G Skill F5-6000J3038F16G 2000GB Corsair MP700 PRO AMD Radeon RX 7900 GRE 16GB AMD Navi 31 HDMI/DP DELL U2723QE Intel I225-V + Intel Wi-Fi 6E Ubuntu 24.04 6.10.0-phx (x86_64) GNOME Shell 46.0 X Server + Wayland 4.6 Mesa 24.2~git2406040600.8112d4~oibaf~n (git-8112d44 2024-06-04 noble-oibaf-ppa) (LLVM 17.0.6 DRM 3.57) GCC 13.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/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-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: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xb40401c Python Details - Python 3.12.3 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 + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: 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 / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
9600X stockfish: Chess Benchmark graphics-magick: Resizing z3: 1.smt2 lczero: Eigen lczero: BLAS graphics-magick: Rotate mnn: mobilenetV3 simdjson: DistinctUserID onnx: ZFNet-512 - CPU - Parallel z3: 2.smt2 onnx: ArcFace ResNet-100 - CPU - Parallel mnn: inception-v3 mnn: squeezenetv1.1 onnx: GPT-2 - CPU - Parallel graphics-magick: HWB Color Space onnx: T5 Encoder - CPU - Parallel x265: Bosphorus 4K graphics-magick: Noise-Gaussian xnnpack: FP32MobileNetV3Large y-cruncher: 500M onnx: ResNet50 v1-12-int8 - CPU - Parallel x265: Bosphorus 1080p mnn: SqueezeNetV1.0 onnx: fcn-resnet101-11 - CPU - Parallel onnx: bertsquad-12 - CPU - Parallel onnx: T5 Encoder - CPU - Standard onnx: ZFNet-512 - CPU - Standard onnx: yolov4 - CPU - Standard y-cruncher: 1B mnn: resnet-v2-50 onnx: yolov4 - CPU - Parallel onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard xnnpack: QU8MobileNetV2 xnnpack: FP16MobileNetV3Small stockfish: Chess Benchmark graphics-magick: Swirl simdjson: Kostya svt-av1: Preset 5 - Bosphorus 4K etcpak: Multi-Threaded - ETC2 simdjson: TopTweet blender: BMW27 - CPU-Only blender: Fishy Cat - CPU-Only onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel svt-av1: Preset 8 - Beauty 4K 10-bit onnx: CaffeNet 12-int8 - CPU - Parallel onnx: super-resolution-10 - CPU - Standard onnx: super-resolution-10 - CPU - Parallel svt-av1: Preset 8 - Bosphorus 4K onnx: bertsquad-12 - CPU - Standard ospray: particle_volume/pathtracer/real_time onnx: ResNet50 v1-12-int8 - CPU - Standard blender: Junkshop - CPU-Only compress-lz4: 9 - Decompression Speed simdjson: LargeRand onnx: CaffeNet 12-int8 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Standard svt-av1: Preset 8 - Bosphorus 1080p xnnpack: FP16MobileNetV3Large gromacs: water_GMX50_bare compress-lz4: 1 - Compression Speed xnnpack: FP32MobileNetV2 byte: Pipe mt-dgemm: Sustained Floating-Point Rate xnnpack: FP32MobileNetV3Small mnn: mobilenet-v1-1.0 svt-av1: Preset 5 - Beauty 4K 10-bit compress-lz4: 2 - Compression Speed mnn: MobileNetV2_224 onnx: ResNet101_DUC_HDC-12 - CPU - Standard ospray: gravity_spheres_volume/dim_512/pathtracer/real_time ospray: particle_volume/ao/real_time onnx: ResNet101_DUC_HDC-12 - CPU - Parallel build2: Time To Compile onnx: GPT-2 - CPU - Standard ospray: gravity_spheres_volume/dim_512/ao/real_time svt-av1: Preset 3 - Bosphorus 4K svt-av1: Preset 13 - Bosphorus 4K simdjson: PartialTweets ospray: gravity_spheres_volume/dim_512/scivis/real_time svt-av1: Preset 5 - Bosphorus 1080p compress-lz4: 12 - Compression Speed compress-lz4: 1 - Decompression Speed xnnpack: QU8MobileNetV3Large blender: Classroom - CPU-Only svt-av1: Preset 3 - Bosphorus 1080p byte: Dhrystone 2 blender: Pabellon Barcelona - CPU-Only whisperfile: Medium svt-av1: Preset 13 - Beauty 4K 10-bit svt-av1: Preset 3 - Beauty 4K 10-bit mnn: nasnet gromacs: MPI CPU - water_GMX50_bare byte: Whetstone Double compress-lz4: 3 - Decompression Speed xnnpack: FP16MobileNetV2 onnx: ArcFace ResNet-100 - CPU - Standard ospray: particle_volume/scivis/real_time byte: System Call svt-av1: Preset 13 - Bosphorus 1080p whisperfile: Small povray: Trace Time blender: Barbershop - CPU-Only compress-lz4: 9 - Compression Speed compress-lz4: 2 - Decompression Speed compress-lz4: 3 - Compression Speed whisperfile: Tiny compress-lz4: 12 - Decompression Speed xnnpack: QU8MobileNetV3Small oidn: RTLightmap.hdr.4096x4096 - CPU-Only oidn: RT.ldr_alb_nrm.3840x2160 - CPU-Only oidn: RT.hdr_alb_nrm.3840x2160 - CPU-Only graphics-magick: Enhanced graphics-magick: Sharpen onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel onnx: ResNet101_DUC_HDC-12 - CPU - Standard onnx: ResNet101_DUC_HDC-12 - CPU - Parallel onnx: super-resolution-10 - CPU - Standard onnx: super-resolution-10 - CPU - Parallel onnx: ResNet50 v1-12-int8 - CPU - Standard onnx: ResNet50 v1-12-int8 - CPU - Parallel onnx: ArcFace ResNet-100 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Parallel onnx: fcn-resnet101-11 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Parallel onnx: CaffeNet 12-int8 - CPU - Standard onnx: CaffeNet 12-int8 - CPU - Parallel onnx: bertsquad-12 - CPU - Standard onnx: bertsquad-12 - CPU - Parallel onnx: T5 Encoder - CPU - Standard onnx: T5 Encoder - CPU - Parallel onnx: ZFNet-512 - CPU - Standard onnx: ZFNet-512 - CPU - Parallel onnx: yolov4 - CPU - Standard onnx: yolov4 - CPU - Parallel onnx: GPT-2 - CPU - Standard onnx: GPT-2 - CPU - Parallel a b 26172950 283 16.403 296 395 212 0.587 13.37 54.971 36.985 23.4038 11.166 0.95 51.751 257 66.1263 18.3 68 1040 10.088 131.248 96 1.491 1.61485 12.718 163.537 113.38 15.1397 23.42 6.904 8.4957 59.5044 589 478 20554668 126 7.58 21.896 291.957 13.1 113.2 159.34 39.058 5.113 173.531 205.487 67.8881 50.953 26.9653 184.845 536.301 159.46 6446.9 2.06 869.325 2.80898 171.004 1198 1.06 1033.44 846 44538458.4 144.483777 301 1.218 3.765 474.96 1.079 1.30568 4.37639 4.74195 0.866341 163.78 140.537 3.83541 5.597 167.003 12.42 3.74914 71.958 20.74 6687.6 604 317.03 19.802 781175751 357.14 424.17256 8.898 0.869 4.357 1.741 149214.7 6167.8 1252 52.079 4.75488 43927689.4 706.48 156.75258 38.055 1128.79 60.49 5923.8 168.65 32.88129 6644.3 290 0.33 0.69 0.69 50 19 16.8024 25.6007 765.883 1154.28 4.86562 14.7293 1.8639 7.61827 19.1986 42.7267 355.998 619.25 1.14962 5.76164 37.0821 78.6265 6.11301 15.1213 8.81821 18.189 66.0492 117.704 7.11087 19.32 21488082 235 17.811 277 372 200 0.556 12.73 57.1637 38.081 24.0812 10.862 0.926 52.8859 252 64.8427 18.6 67 1025 9.946 132.978 97.25 1.51 1.63528 12.5592 161.523 114.774 15.3183 23.652 6.837 8.41617 58.9869 594 482 20386602 125 7.52 21.724 294.198 13.2 112.43 160.43 38.8065 5.081 172.466 206.734 67.4844 50.654 27.1211 183.844 533.488 160.26 6478.6 2.05 865.224 2.82195 170.251 1193 1.056 1029.67 849 44695866.6 144.97094 302 1.214 3.777 476.47 1.082 1.30931 4.36447 4.72912 0.864063 164.208 140.898 3.84506 5.611 166.587 12.39 3.74053 71.817 20.7 6700.2 605 316.58 19.775 780130318.1 356.68 423.64619 8.887 0.87 4.352 1.743 149060 6174.2 1251 52.038 4.75153 43955081.6 706.919 156.82827 38.072 1129.22 60.47 5922.1 168.61 32.87604 6643.3 290 0.33 0.69 0.69 50 19 16.9498 25.7666 763.756 1157.24 4.83625 14.8173 1.87362 7.51917 19.215 41.5246 354.362 611.513 1.15518 5.79732 36.8685 79.6206 6.18885 15.4208 8.71089 17.4922 65.279 118.816 7.09276 18.9057 OpenBenchmarking.org
Stockfish Chess Benchmark OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 17 Chess Benchmark a b 6M 12M 18M 24M 30M 26172950 21488082 1. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -msse -msse3 -mpopcnt -mavx2 -mbmi -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto-partition=one -flto=jobserver
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Resizing a b 60 120 180 240 300 283 235 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
Z3 Theorem Prover SMT File: 1.smt2 OpenBenchmarking.org Seconds, Fewer Is Better Z3 Theorem Prover 4.12.1 SMT File: 1.smt2 a b 4 8 12 16 20 16.40 17.81 1. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: Eigen a b 60 120 180 240 300 296 277 1. (CXX) g++ options: -flto -pthread
LeelaChessZero Backend: BLAS OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: BLAS a b 90 180 270 360 450 395 372 1. (CXX) g++ options: -flto -pthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Rotate a b 50 100 150 200 250 212 200 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
Mobile Neural Network Model: mobilenetV3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: mobilenetV3 a b 0.1321 0.2642 0.3963 0.5284 0.6605 0.587 0.556 MIN: 0.55 / MAX: 1.28 MIN: 0.53 / MAX: 1.25 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
simdjson Throughput Test: DistinctUserID OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: DistinctUserID a b 3 6 9 12 15 13.37 12.73 1. (CXX) g++ options: -O3 -lrt
ONNX Runtime Model: ZFNet-512 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Parallel a b 13 26 39 52 65 54.97 57.16 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
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 9 18 27 36 45 36.99 38.08 1. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel a b 6 12 18 24 30 23.40 24.08 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Mobile Neural Network Model: inception-v3 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: inception-v3 a b 3 6 9 12 15 11.17 10.86 MIN: 10.11 / MAX: 20.26 MIN: 8.94 / MAX: 20.25 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
Mobile Neural Network Model: squeezenetv1.1 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: squeezenetv1.1 a b 0.2138 0.4276 0.6414 0.8552 1.069 0.950 0.926 MIN: 0.87 / MAX: 1.67 MIN: 0.87 / MAX: 1.74 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Parallel a b 12 24 36 48 60 51.75 52.89 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: HWB Color Space a b 60 120 180 240 300 257 252 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
ONNX Runtime Model: T5 Encoder - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Parallel a b 15 30 45 60 75 66.13 64.84 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
x265 Video Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better x265 Video Input: Bosphorus 4K a b 5 10 15 20 25 18.3 18.6 1. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Noise-Gaussian a b 15 30 45 60 75 68 67 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
XNNPACK Model: FP32MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Large a b 200 400 600 800 1000 1040 1025 1. (CXX) g++ options: -O3 -lrt -lm
Y-Cruncher Pi Digits To Calculate: 500M OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.5 Pi Digits To Calculate: 500M a b 3 6 9 12 15 10.088 9.946
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel a b 30 60 90 120 150 131.25 132.98 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
x265 Video Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better x265 Video Input: Bosphorus 1080p a b 20 40 60 80 100 96.00 97.25 1. x265 [info]: HEVC encoder version 3.5+1-f0c1022b6
Mobile Neural Network Model: SqueezeNetV1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: SqueezeNetV1.0 a b 0.3398 0.6796 1.0194 1.3592 1.699 1.491 1.510 MIN: 1.37 / MAX: 2.7 MIN: 1.37 / MAX: 2.63 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel a b 0.3679 0.7358 1.1037 1.4716 1.8395 1.61485 1.63528 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Parallel a b 3 6 9 12 15 12.72 12.56 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: T5 Encoder - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Standard a b 40 80 120 160 200 163.54 161.52 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ZFNet-512 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Standard a b 30 60 90 120 150 113.38 114.77 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Standard a b 4 8 12 16 20 15.14 15.32 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Y-Cruncher Pi Digits To Calculate: 1B OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.8.5 Pi Digits To Calculate: 1B a b 6 12 18 24 30 23.42 23.65
Mobile Neural Network Model: resnet-v2-50 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: resnet-v2-50 a b 2 4 6 8 10 6.904 6.837 MIN: 6.1 / MAX: 16.18 MIN: 6.05 / MAX: 16.69 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Parallel a b 2 4 6 8 10 8.49570 8.41617 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard a b 13 26 39 52 65 59.50 58.99 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
XNNPACK Model: QU8MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV2 a b 130 260 390 520 650 589 594 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Small a b 100 200 300 400 500 478 482 1. (CXX) g++ options: -O3 -lrt -lm
Stockfish Chess Benchmark OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish Chess Benchmark a b 4M 8M 12M 16M 20M 20554668 20386602 1. Stockfish 16 by the Stockfish developers (see AUTHORS file)
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Swirl a b 30 60 90 120 150 126 125 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
simdjson Throughput Test: Kostya OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: Kostya a b 2 4 6 8 10 7.58 7.52 1. (CXX) g++ options: -O3 -lrt
SVT-AV1 Encoder Mode: Preset 5 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Bosphorus 4K a b 5 10 15 20 25 21.90 21.72 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Etcpak Benchmark: Multi-Threaded - Configuration: ETC2 OpenBenchmarking.org Mpx/s, More Is Better Etcpak 2.0 Benchmark: Multi-Threaded - Configuration: ETC2 a b 60 120 180 240 300 291.96 294.20 1. (CXX) g++ options: -flto -pthread
simdjson Throughput Test: TopTweet OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: TopTweet a b 3 6 9 12 15 13.1 13.2 1. (CXX) g++ options: -O3 -lrt
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: BMW27 - Compute: CPU-Only a b 30 60 90 120 150 113.20 112.43
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Fishy Cat - Compute: CPU-Only a b 40 80 120 160 200 159.34 160.43
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel a b 9 18 27 36 45 39.06 38.81 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Encoder Mode: Preset 8 - Input: Beauty 4K 10-bit OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Beauty 4K 10-bit a b 1.1504 2.3008 3.4512 4.6016 5.752 5.113 5.081 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel a b 40 80 120 160 200 173.53 172.47 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Standard a b 50 100 150 200 250 205.49 206.73 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Parallel a b 15 30 45 60 75 67.89 67.48 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Bosphorus 4K a b 11 22 33 44 55 50.95 50.65 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Standard a b 6 12 18 24 30 26.97 27.12 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OSPRay Benchmark: particle_volume/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/pathtracer/real_time a b 40 80 120 160 200 184.85 183.84
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard a b 120 240 360 480 600 536.30 533.49 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Blender Blend File: Junkshop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Junkshop - Compute: CPU-Only a b 40 80 120 160 200 159.46 160.26
LZ4 Compression Compression Level: 9 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 9 - Decompression Speed a b 1400 2800 4200 5600 7000 6446.9 6478.6 1. (CC) gcc options: -O3 -pthread
simdjson Throughput Test: LargeRandom OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: LargeRandom a b 0.4635 0.927 1.3905 1.854 2.3175 2.06 2.05 1. (CXX) g++ options: -O3 -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard a b 200 400 600 800 1000 869.33 865.22 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard a b 0.6349 1.2698 1.9047 2.5396 3.1745 2.80898 2.82195 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 8 - Input: Bosphorus 1080p a b 40 80 120 160 200 171.00 170.25 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
XNNPACK Model: FP16MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Large a b 300 600 900 1200 1500 1198 1193 1. (CXX) g++ options: -O3 -lrt -lm
GROMACS Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS Input: water_GMX50_bare a b 0.2385 0.477 0.7155 0.954 1.1925 1.060 1.056 1. GROMACS version: 2023.3-Ubuntu_2023.3_1ubuntu3
LZ4 Compression Compression Level: 1 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 1 - Compression Speed a b 200 400 600 800 1000 1033.44 1029.67 1. (CC) gcc options: -O3 -pthread
XNNPACK Model: FP32MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV2 a b 200 400 600 800 1000 846 849 1. (CXX) g++ options: -O3 -lrt -lm
BYTE Unix Benchmark Computational Test: Pipe OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 5.1.3-git Computational Test: Pipe a b 10M 20M 30M 40M 50M 44538458.4 44695866.6 1. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm
ACES DGEMM Sustained Floating-Point Rate OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate a b 30 60 90 120 150 144.48 144.97 1. (CC) gcc options: -ffast-math -mavx2 -O3 -fopenmp -lopenblas
XNNPACK Model: FP32MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Small a b 70 140 210 280 350 301 302 1. (CXX) g++ options: -O3 -lrt -lm
Mobile Neural Network Model: mobilenet-v1-1.0 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: mobilenet-v1-1.0 a b 0.2741 0.5482 0.8223 1.0964 1.3705 1.218 1.214 MIN: 1.13 / MAX: 2.19 MIN: 1.14 / MAX: 2.17 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
SVT-AV1 Encoder Mode: Preset 5 - Input: Beauty 4K 10-bit OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Beauty 4K 10-bit a b 0.8498 1.6996 2.5494 3.3992 4.249 3.765 3.777 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
LZ4 Compression Compression Level: 2 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 2 - Compression Speed a b 100 200 300 400 500 474.96 476.47 1. (CC) gcc options: -O3 -pthread
Mobile Neural Network Model: MobileNetV2_224 OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: MobileNetV2_224 a b 0.2435 0.487 0.7305 0.974 1.2175 1.079 1.082 MIN: 0.98 / MAX: 4.28 MIN: 0.97 / MAX: 2.17 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
ONNX Runtime Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard a b 0.2946 0.5892 0.8838 1.1784 1.473 1.30568 1.30931 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OSPRay Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time a b 0.9847 1.9694 2.9541 3.9388 4.9235 4.37639 4.36447
OSPRay Benchmark: particle_volume/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/ao/real_time a b 1.0669 2.1338 3.2007 4.2676 5.3345 4.74195 4.72912
ONNX Runtime Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel a b 0.1949 0.3898 0.5847 0.7796 0.9745 0.866341 0.864063 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Build2 Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.17 Time To Compile a b 40 80 120 160 200 163.78 164.21
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Standard a b 30 60 90 120 150 140.54 140.90 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OSPRay Benchmark: gravity_spheres_volume/dim_512/ao/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: gravity_spheres_volume/dim_512/ao/real_time a b 0.8651 1.7302 2.5953 3.4604 4.3255 3.83541 3.84506
SVT-AV1 Encoder Mode: Preset 3 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Bosphorus 4K a b 1.2625 2.525 3.7875 5.05 6.3125 5.597 5.611 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Bosphorus 4K a b 40 80 120 160 200 167.00 166.59 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
simdjson Throughput Test: PartialTweets OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: PartialTweets a b 3 6 9 12 15 12.42 12.39 1. (CXX) g++ options: -O3 -lrt
OSPRay Benchmark: gravity_spheres_volume/dim_512/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time a b 0.8436 1.6872 2.5308 3.3744 4.218 3.74914 3.74053
SVT-AV1 Encoder Mode: Preset 5 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 5 - Input: Bosphorus 1080p a b 16 32 48 64 80 71.96 71.82 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
LZ4 Compression Compression Level: 12 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 12 - Compression Speed a b 5 10 15 20 25 20.74 20.70 1. (CC) gcc options: -O3 -pthread
LZ4 Compression Compression Level: 1 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 1 - Decompression Speed a b 1400 2800 4200 5600 7000 6687.6 6700.2 1. (CC) gcc options: -O3 -pthread
XNNPACK Model: QU8MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Large a b 130 260 390 520 650 604 605 1. (CXX) g++ options: -O3 -lrt -lm
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Classroom - Compute: CPU-Only a b 70 140 210 280 350 317.03 316.58
SVT-AV1 Encoder Mode: Preset 3 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Bosphorus 1080p a b 5 10 15 20 25 19.80 19.78 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
BYTE Unix Benchmark Computational Test: Dhrystone 2 OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 5.1.3-git Computational Test: Dhrystone 2 a b 200M 400M 600M 800M 1000M 781175751.0 780130318.1 1. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Pabellon Barcelona - Compute: CPU-Only a b 80 160 240 320 400 357.14 356.68
Whisperfile Model Size: Medium OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Medium a b 90 180 270 360 450 424.17 423.65
SVT-AV1 Encoder Mode: Preset 13 - Input: Beauty 4K 10-bit OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Beauty 4K 10-bit a b 2 4 6 8 10 8.898 8.887 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
SVT-AV1 Encoder Mode: Preset 3 - Input: Beauty 4K 10-bit OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 3 - Input: Beauty 4K 10-bit a b 0.1958 0.3916 0.5874 0.7832 0.979 0.869 0.870 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Mobile Neural Network Model: nasnet OpenBenchmarking.org ms, Fewer Is Better Mobile Neural Network 2.9.b11b7037d Model: nasnet a b 0.9803 1.9606 2.9409 3.9212 4.9015 4.357 4.352 MIN: 3.81 / MAX: 13.67 MIN: 3.84 / MAX: 9.48 1. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -pthread -ldl
GROMACS Implementation: MPI CPU - Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare a b 0.3922 0.7844 1.1766 1.5688 1.961 1.741 1.743 1. (CXX) g++ options: -O3 -lm
BYTE Unix Benchmark Computational Test: Whetstone Double OpenBenchmarking.org MWIPS, More Is Better BYTE Unix Benchmark 5.1.3-git Computational Test: Whetstone Double a b 30K 60K 90K 120K 150K 149214.7 149060.0 1. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm
LZ4 Compression Compression Level: 3 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 3 - Decompression Speed a b 1300 2600 3900 5200 6500 6167.8 6174.2 1. (CC) gcc options: -O3 -pthread
XNNPACK Model: FP16MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV2 a b 300 600 900 1200 1500 1252 1251 1. (CXX) g++ options: -O3 -lrt -lm
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard a b 12 24 36 48 60 52.08 52.04 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
OSPRay Benchmark: particle_volume/scivis/real_time OpenBenchmarking.org Items Per Second, More Is Better OSPRay 3.2 Benchmark: particle_volume/scivis/real_time a b 1.0698 2.1396 3.2094 4.2792 5.349 4.75488 4.75153
BYTE Unix Benchmark Computational Test: System Call OpenBenchmarking.org LPS, More Is Better BYTE Unix Benchmark 5.1.3-git Computational Test: System Call a b 9M 18M 27M 36M 45M 43927689.4 43955081.6 1. (CC) gcc options: -pedantic -O3 -ffast-math -march=native -mtune=native -lm
SVT-AV1 Encoder Mode: Preset 13 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2.2 Encoder Mode: Preset 13 - Input: Bosphorus 1080p a b 150 300 450 600 750 706.48 706.92 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
Whisperfile Model Size: Small OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Small a b 30 60 90 120 150 156.75 156.83
POV-Ray Trace Time OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray Trace Time a b 9 18 27 36 45 38.06 38.07 1. POV-Ray 3.7.0.10.unofficial
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.2 Blend File: Barbershop - Compute: CPU-Only a b 200 400 600 800 1000 1128.79 1129.22
LZ4 Compression Compression Level: 9 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 9 - Compression Speed a b 14 28 42 56 70 60.49 60.47 1. (CC) gcc options: -O3 -pthread
LZ4 Compression Compression Level: 2 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 2 - Decompression Speed a b 1300 2600 3900 5200 6500 5923.8 5922.1 1. (CC) gcc options: -O3 -pthread
LZ4 Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 3 - Compression Speed a b 40 80 120 160 200 168.65 168.61 1. (CC) gcc options: -O3 -pthread
Whisperfile Model Size: Tiny OpenBenchmarking.org Seconds, Fewer Is Better Whisperfile 20Aug24 Model Size: Tiny a b 8 16 24 32 40 32.88 32.88
LZ4 Compression Compression Level: 12 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.10 Compression Level: 12 - Decompression Speed a b 1400 2800 4200 5600 7000 6644.3 6643.3 1. (CC) gcc options: -O3 -pthread
XNNPACK Model: QU8MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Small a b 60 120 180 240 300 290 290 1. (CXX) g++ options: -O3 -lrt -lm
Intel Open Image Denoise Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.3 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only a b 0.0743 0.1486 0.2229 0.2972 0.3715 0.33 0.33
Intel Open Image Denoise Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.3 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only a b 0.1553 0.3106 0.4659 0.6212 0.7765 0.69 0.69
Intel Open Image Denoise Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 2.3 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only a b 0.1553 0.3106 0.4659 0.6212 0.7765 0.69 0.69
GraphicsMagick Operation: Enhanced OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Enhanced a b 11 22 33 44 55 50 50 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick Operation: Sharpen a b 5 10 15 20 25 19 19 1. GraphicsMagick 1.3.42 2023-09-23 Q16 http://www.GraphicsMagick.org/
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard a b 4 8 12 16 20 16.80 16.95 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel a b 6 12 18 24 30 25.60 25.77 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Standard a b 170 340 510 680 850 765.88 763.76 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet101_DUC_HDC-12 - Device: CPU - Executor: Parallel a b 200 400 600 800 1000 1154.28 1157.24 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Standard a b 1.0948 2.1896 3.2844 4.3792 5.474 4.86562 4.83625 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: super-resolution-10 - Device: CPU - Executor: Parallel a b 4 8 12 16 20 14.73 14.82 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard a b 0.4216 0.8432 1.2648 1.6864 2.108 1.86390 1.87362 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel a b 2 4 6 8 10 7.61827 7.51917 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard a b 5 10 15 20 25 19.20 19.22 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel a b 10 20 30 40 50 42.73 41.52 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard a b 80 160 240 320 400 356.00 354.36 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel a b 130 260 390 520 650 619.25 611.51 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard a b 0.2599 0.5198 0.7797 1.0396 1.2995 1.14962 1.15518 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel a b 1.3044 2.6088 3.9132 5.2176 6.522 5.76164 5.79732 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Standard a b 9 18 27 36 45 37.08 36.87 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: bertsquad-12 - Device: CPU - Executor: Parallel a b 20 40 60 80 100 78.63 79.62 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: T5 Encoder - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Standard a b 2 4 6 8 10 6.11301 6.18885 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: T5 Encoder - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: T5 Encoder - Device: CPU - Executor: Parallel a b 4 8 12 16 20 15.12 15.42 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ZFNet-512 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Standard a b 2 4 6 8 10 8.81821 8.71089 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ZFNet-512 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: ZFNet-512 - Device: CPU - Executor: Parallel a b 4 8 12 16 20 18.19 17.49 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Standard a b 15 30 45 60 75 66.05 65.28 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: yolov4 - Device: CPU - Executor: Parallel a b 30 60 90 120 150 117.70 118.82 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Standard a b 2 4 6 8 10 7.11087 7.09276 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.19 Model: GPT-2 - Device: CPU - Executor: Parallel a b 5 10 15 20 25 19.32 18.91 1. (CXX) g++ options: -O3 -march=native -ffunction-sections -fdata-sections -mtune=native -flto=auto -fno-fat-lto-objects -ldl -lrt
Phoronix Test Suite v10.8.5