Desktop machine learning AMD Ryzen 9 3900X 12-Core testing with a MSI X570-A PRO (MS-7C37) v3.0 (H.70 BIOS) and NVIDIA GeForce RTX 3060 12GB on Ubuntu 23.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2405015-VPA1-DESKTOP46&sor&grs .
Desktop machine learning Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Display Server Display Driver OpenCL Compiler File-System Screen Resolution mantic mantic-no-omit-framepointer noble AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads) MSI X570-A PRO (MS-7C37) v3.0 (H.70 BIOS) AMD Starship/Matisse 2 x 16GB DDR4-3200MT/s F4-3200C16-16GVK 2000GB Seagate ST2000DM006-2DM1 + 2000GB Western Digital WD20EZAZ-00G + 500GB Samsung SSD 860 + 8002GB Seagate ST8000DM004-2CX1 + 1000GB CT1000BX500SSD1 + 512GB TS512GESD310C NVIDIA GeForce RTX 3060 12GB NVIDIA GA104 HD Audio DELL P2314H Realtek RTL8111/8168/8411 Ubuntu 23.10 6.5.0-9-generic (x86_64) X Server 1.21.1.7 NVIDIA OpenCL 3.0 CUDA 12.2.146 GCC 13.2.0 + CUDA 12.2 ext4 1920x1080 NVIDIA GeForce RTX 3060 DELL P2314H + U32J59x Realtek RTL8111/8168/8211/8411 Ubuntu 24.04 6.8.0-31-generic (x86_64) GCC 13.2.0 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - mantic: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - mantic-no-omit-framepointer: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-b9QCDx/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-b9QCDx/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - noble: --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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8701013 Python Details - mantic: Python 3.11.6 - mantic-no-omit-framepointer: Python 3.11.6 - noble: Python 3.12.3 Security Details - mantic: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - mantic-no-omit-framepointer: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - noble: 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: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected Environment Details - mantic-no-omit-framepointer: CXXFLAGS=-fno-omit-frame-pointer QMAKE_CFLAGS=-fno-omit-frame-pointer CFLAGS=-fno-omit-frame-pointer CFLAGS_OVERRIDE=-fno-omit-frame-pointer QMAKE_CXXFLAGS=-fno-omit-frame-pointer FFLAGS=-fno-omit-frame-pointer - noble: CXXFLAGS="-fno-omit-frame-pointer -frecord-gcc-switches -O2" QMAKE_CFLAGS="-fno-omit-frame-pointer -frecord-gcc-switches -O2" CFLAGS="-fno-omit-frame-pointer -frecord-gcc-switches -O2" CFLAGS_OVERRIDE="-fno-omit-frame-pointer -frecord-gcc-switches -O2" QMAKE_CXXFLAGS="-fno-omit-frame-pointer -frecord-gcc-switches -O2" FFLAGS="-fno-omit-frame-pointer -frecord-gcc-switches -O2"
Desktop machine learning scikit-learn: Lasso scikit-learn: SGD Regression scikit-learn: Plot OMP vs. LARS scikit-learn: TSNE MNIST Dataset pyperformance: json_loads pyperformance: python_startup scikit-learn: Isotonic / Logistic scikit-learn: Sample Without Replacement scikit-learn: Tree pyhpc: GPU - Numpy - 16384 - Isoneutral Mixing pyhpc: CPU - Numpy - 16384 - Isoneutral Mixing scikit-learn: Isotonic / Perturbed Logarithm scikit-learn: GLM scikit-learn: Text Vectorizers scikit-learn: Hist Gradient Boosting Adult pybench: Total For Average Test Times pyperformance: go scikit-learn: Sparse Rand Projections / 100 Iterations pytorch: NVIDIA CUDA GPU - 16 - Efficientnet_v2_l scikit-learn: Hist Gradient Boosting Categorical Only pyhpc: GPU - Numpy - 262144 - Equation of State scikit-learn: Hist Gradient Boosting pyhpc: CPU - Numpy - 65536 - Equation of State pyhpc: GPU - Numpy - 262144 - Isoneutral Mixing scikit-learn: LocalOutlierFactor pytorch: NVIDIA CUDA GPU - 1 - Efficientnet_v2_l pyhpc: CPU - Numpy - 262144 - Equation of State pyperformance: raytrace scikit-learn: Kernel PCA Solvers / Time vs. N Samples scikit-learn: Plot Neighbors scikit-learn: Plot Polynomial Kernel Approximation pyhpc: CPU - Numpy - 4194304 - Isoneutral Mixing pyperformance: django_template pyperformance: regex_compile pyhpc: CPU - Numpy - 65536 - Isoneutral Mixing pyhpc: GPU - Numpy - 65536 - Isoneutral Mixing scikit-learn: Plot Ward scikit-learn: Covertype Dataset Benchmark pytorch: NVIDIA CUDA GPU - 64 - ResNet-152 pyperformance: pathlib pyhpc: GPU - Numpy - 4194304 - Equation of State pyhpc: CPU - Numpy - 4194304 - Equation of State pyperformance: crypto_pyaes pytorch: NVIDIA CUDA GPU - 1 - ResNet-152 pytorch: NVIDIA CUDA GPU - 64 - ResNet-50 scikit-learn: Kernel PCA Solvers / Time vs. N Components pyhpc: CPU - Numpy - 1048576 - Isoneutral Mixing pytorch: NVIDIA CUDA GPU - 256 - Efficientnet_v2_l scikit-learn: Plot Hierarchical pytorch: NVIDIA CUDA GPU - 512 - ResNet-152 pytorch: CPU - 32 - ResNet-152 pyhpc: GPU - Numpy - 4194304 - Isoneutral Mixing scikit-learn: Sparsify pytorch: NVIDIA CUDA GPU - 64 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 256 - ResNet-152 pytorch: NVIDIA CUDA GPU - 32 - ResNet-50 pyperformance: pickle_pure_python pyhpc: CPU - Numpy - 262144 - Isoneutral Mixing scikit-learn: SGDOneClassSVM pytorch: NVIDIA CUDA GPU - 32 - Efficientnet_v2_l pyhpc: GPU - Numpy - 1048576 - Isoneutral Mixing scikit-learn: Plot Incremental PCA pytorch: CPU - 256 - ResNet-152 scikit-learn: Feature Expansions pyperformance: 2to3 pytorch: CPU - 1 - ResNet-152 pyperformance: chaos scikit-learn: Hist Gradient Boosting Threading pyperformance: nbody pyhpc: GPU - Numpy - 1048576 - Equation of State pytorch: NVIDIA CUDA GPU - 32 - ResNet-152 numpy: pytorch: NVIDIA CUDA GPU - 16 - ResNet-152 pytorch: NVIDIA CUDA GPU - 512 - ResNet-50 pytorch: CPU - 32 - ResNet-50 scikit-learn: 20 Newsgroups / Logistic Regression pytorch: CPU - 32 - Efficientnet_v2_l pytorch: CPU - 16 - Efficientnet_v2_l pytorch: CPU - 512 - Efficientnet_v2_l pytorch: CPU - 64 - Efficientnet_v2_l pytorch: CPU - 64 - ResNet-50 pyhpc: CPU - Numpy - 1048576 - Equation of State pyperformance: float pytorch: CPU - 512 - ResNet-152 scikit-learn: MNIST Dataset pytorch: CPU - 512 - ResNet-50 scikit-learn: SAGA pytorch: CPU - 1 - ResNet-50 pytorch: CPU - 16 - ResNet-50 pytorch: NVIDIA CUDA GPU - 512 - Efficientnet_v2_l pytorch: CPU - 256 - Efficientnet_v2_l pytorch: CPU - 16 - ResNet-152 pytorch: CPU - 64 - ResNet-152 pytorch: CPU - 256 - ResNet-50 pytorch: NVIDIA CUDA GPU - 1 - ResNet-50 pytorch: NVIDIA CUDA GPU - 256 - ResNet-50 pytorch: CPU - 1 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 16 - ResNet-50 pyhpc: GPU - Numpy - 65536 - Equation of State pyhpc: CPU - Numpy - 16384 - Equation of State scikit-learn: Isolation Forest pyhpc: GPU - Numpy - 16384 - Equation of State mantic mantic-no-omit-framepointer noble 511.848 106.315 91.499 236.865 19.5 7.61 1470.806 158.262 48.338 0.009 0.009 1788.259 293.598 60.814 103.497 774 129 613.547 38.95 18.579 0.062 109.984 0.015 0.131 53.464 39.35 0.061 262 72.541 147.752 150.732 2.670 28.5 116 0.032 0.033 57.824 376.145 71.81 19.7 1.422 1.402 65.1 73.91 201.41 37.242 0.619 37.36 211.286 72.31 9.84 2.662 127.282 37.88 71.74 199.46 259 0.131 379.739 37.71 0.631 31.006 9.77 131.277 221 12.72 62.8 110.215 76.2 0.263 74.15 426.28 73.01 203.18 24.29 41.519 5.63 5.63 5.61 5.62 24.24 0.263 67.4 9.87 65.763 24.13 868.018 32.36 24.28 37.43 5.61 9.88 9.88 24.42 210.88 202.72 7.31 200.30 0.015 0.003 289.371 0.003 509.537 107.527 92.582 236.786 20.8 7.64 1471.834 161.460 52.969 0.008 0.008 1828.300 295.096 63.875 105.647 790 131 631.071 36.10 18.865 0.058 111.255 0.015 0.128 56.754 37.29 0.058 274 72.909 142.451 150.376 2.626 29.5 120 0.032 0.033 57.545 370.694 73.65 20.2 1.411 1.405 66.6 72.27 205.95 37.889 0.618 36.60 208.391 73.75 10.00 2.620 125.442 37.24 72.91 202.68 263 0.132 382.611 37.16 0.622 31.057 9.91 133.092 224 12.78 63.6 110.374 77.1 0.260 73.36 428.61 72.24 201.14 24.35 41.728 5.64 5.64 5.65 5.65 24.40 0.262 66.9 9.80 65.877 24.28 873.822 32.54 24.38 37.22 5.64 9.93 9.91 24.37 211.46 203.22 7.32 200.17 0.015 0.003 336.372 0.002 345.400 78.880 68.172 285.823 22.8 8.76 1684.546 179.638 47.033 0.008 0.009 1963.772 269.806 66.393 112.713 839 121 663.953 19.932 0.061 117.407 0.016 0.136 54.288 0.060 70.022 142.159 145.363 2.720 0.033 0.034 56.132 381.447 1.446 1.436 37.107 0.631 207.104 9.81 2.668 125.069 0.133 385.383 0.630 30.617 9.86 133.144 12.89 111.554 0.262 430.83 24.12 41.914 5.59 5.59 5.60 5.60 24.19 0.261 9.87 65.416 24.30 869.369 32.34 24.43 5.61 9.88 9.87 24.33 7.31 0.015 0.003 314.034 0.003 OpenBenchmarking.org
Scikit-Learn Benchmark: Lasso OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Lasso noble mantic-no-omit-framepointer mantic 110 220 330 440 550 SE +/- 1.37, N = 3 SE +/- 3.50, N = 3 SE +/- 3.22, N = 3 345.40 509.54 511.85 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: SGD Regression OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: SGD Regression noble mantic mantic-no-omit-framepointer 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 1.06, N = 6 SE +/- 0.49, N = 3 78.88 106.32 107.53 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Plot OMP vs. LARS OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot OMP vs. LARS noble mantic mantic-no-omit-framepointer 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.44, N = 3 68.17 91.50 92.58 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: TSNE MNIST Dataset OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: TSNE MNIST Dataset mantic-no-omit-framepointer mantic noble 60 120 180 240 300 SE +/- 0.54, N = 3 SE +/- 0.44, N = 3 SE +/- 0.91, N = 3 236.79 236.87 285.82 -O2 1. (F9X) gfortran options: -O0
PyPerformance Benchmark: json_loads OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: json_loads mantic mantic-no-omit-framepointer noble 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 19.5 20.8 22.8
PyPerformance Benchmark: python_startup OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: python_startup mantic mantic-no-omit-framepointer noble 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 7.61 7.64 8.76
Scikit-Learn Benchmark: Isotonic / Logistic OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Isotonic / Logistic mantic mantic-no-omit-framepointer noble 400 800 1200 1600 2000 SE +/- 12.29, N = 3 SE +/- 14.46, N = 3 SE +/- 9.43, N = 3 1470.81 1471.83 1684.55 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Sample Without Replacement OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Sample Without Replacement mantic mantic-no-omit-framepointer noble 40 80 120 160 200 SE +/- 0.60, N = 3 SE +/- 0.62, N = 3 SE +/- 2.21, N = 3 158.26 161.46 179.64 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Tree OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Tree noble mantic mantic-no-omit-framepointer 12 24 36 48 60 SE +/- 0.52, N = 3 SE +/- 0.59, N = 4 SE +/- 0.48, N = 15 47.03 48.34 52.97 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 16384 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 16384 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer noble mantic 0.002 0.004 0.006 0.008 0.01 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.008 0.008 0.009
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer mantic noble 0.002 0.004 0.006 0.008 0.01 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.008 0.009 0.009
Scikit-Learn Benchmark: Isotonic / Perturbed Logarithm OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Isotonic / Perturbed Logarithm mantic mantic-no-omit-framepointer noble 400 800 1200 1600 2000 SE +/- 24.41, N = 3 SE +/- 16.46, N = 3 SE +/- 1.48, N = 3 1788.26 1828.30 1963.77 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: GLM OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: GLM noble mantic mantic-no-omit-framepointer 60 120 180 240 300 SE +/- 0.93, N = 3 SE +/- 1.06, N = 3 SE +/- 1.07, N = 3 269.81 293.60 295.10 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Text Vectorizers OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Text Vectorizers mantic mantic-no-omit-framepointer noble 15 30 45 60 75 SE +/- 0.19, N = 3 SE +/- 0.08, N = 3 SE +/- 0.32, N = 3 60.81 63.88 66.39 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Hist Gradient Boosting Adult OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Hist Gradient Boosting Adult mantic mantic-no-omit-framepointer noble 30 60 90 120 150 SE +/- 0.70, N = 3 SE +/- 0.59, N = 3 SE +/- 0.52, N = 3 103.50 105.65 112.71 -O2 1. (F9X) gfortran options: -O0
PyBench Total For Average Test Times OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times mantic mantic-no-omit-framepointer noble 200 400 600 800 1000 SE +/- 1.00, N = 3 SE +/- 1.20, N = 3 SE +/- 8.70, N = 4 774 790 839
PyPerformance Benchmark: go OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: go noble mantic mantic-no-omit-framepointer 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 121 129 131
Scikit-Learn Benchmark: Sparse Random Projections / 100 Iterations OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Sparse Random Projections / 100 Iterations mantic mantic-no-omit-framepointer noble 140 280 420 560 700 SE +/- 3.80, N = 3 SE +/- 7.06, N = 4 SE +/- 4.34, N = 3 613.55 631.07 663.95 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 38.95 36.10 MIN: 37.12 / MAX: 39.27 MIN: 34.25 / MAX: 38.01
Scikit-Learn Benchmark: Hist Gradient Boosting Categorical Only OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Hist Gradient Boosting Categorical Only mantic mantic-no-omit-framepointer noble 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 18.58 18.87 19.93 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 262144 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 262144 - Benchmark: Equation of State mantic-no-omit-framepointer noble mantic 0.014 0.028 0.042 0.056 0.07 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.058 0.061 0.062
Scikit-Learn Benchmark: Hist Gradient Boosting OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Hist Gradient Boosting mantic mantic-no-omit-framepointer noble 30 60 90 120 150 SE +/- 0.22, N = 3 SE +/- 0.25, N = 3 SE +/- 0.17, N = 3 109.98 111.26 117.41 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Equation of State mantic mantic-no-omit-framepointer noble 0.0036 0.0072 0.0108 0.0144 0.018 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 15 0.015 0.015 0.016
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 262144 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 262144 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer mantic noble 0.0306 0.0612 0.0918 0.1224 0.153 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.128 0.131 0.136
Scikit-Learn Benchmark: LocalOutlierFactor OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: LocalOutlierFactor mantic noble mantic-no-omit-framepointer 13 26 39 52 65 SE +/- 0.18, N = 3 SE +/- 0.02, N = 3 SE +/- 0.74, N = 15 53.46 54.29 56.75 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.47, N = 3 SE +/- 0.26, N = 3 39.35 37.29 MIN: 36.65 / MAX: 40.42 MIN: 35.83 / MAX: 39.17
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Equation of State mantic-no-omit-framepointer noble mantic 0.0137 0.0274 0.0411 0.0548 0.0685 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.058 0.060 0.061
PyPerformance Benchmark: raytrace OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: raytrace mantic mantic-no-omit-framepointer 60 120 180 240 300 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 262 274
Scikit-Learn Benchmark: Kernel PCA Solvers / Time vs. N Samples OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Kernel PCA Solvers / Time vs. N Samples noble mantic mantic-no-omit-framepointer 16 32 48 64 80 SE +/- 0.44, N = 3 SE +/- 0.05, N = 3 SE +/- 0.16, N = 3 70.02 72.54 72.91 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Plot Neighbors OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot Neighbors noble mantic-no-omit-framepointer mantic 30 60 90 120 150 SE +/- 1.09, N = 3 SE +/- 0.59, N = 3 SE +/- 1.34, N = 7 142.16 142.45 147.75 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Plot Polynomial Kernel Approximation OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot Polynomial Kernel Approximation noble mantic-no-omit-framepointer mantic 30 60 90 120 150 SE +/- 1.46, N = 3 SE +/- 1.20, N = 3 SE +/- 1.22, N = 3 145.36 150.38 150.73 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer mantic noble 0.612 1.224 1.836 2.448 3.06 SE +/- 0.002, N = 3 SE +/- 0.010, N = 3 SE +/- 0.010, N = 3 2.626 2.670 2.720
PyPerformance Benchmark: django_template OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template mantic mantic-no-omit-framepointer 7 14 21 28 35 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 28.5 29.5
PyPerformance Benchmark: regex_compile OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile mantic mantic-no-omit-framepointer 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 116 120
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 65536 - Benchmark: Isoneutral Mixing mantic mantic-no-omit-framepointer noble 0.0074 0.0148 0.0222 0.0296 0.037 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.032 0.032 0.033
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 65536 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 65536 - Benchmark: Isoneutral Mixing mantic mantic-no-omit-framepointer noble 0.0077 0.0154 0.0231 0.0308 0.0385 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.033 0.033 0.034
Scikit-Learn Benchmark: Plot Ward OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot Ward noble mantic-no-omit-framepointer mantic 13 26 39 52 65 SE +/- 0.20, N = 3 SE +/- 0.22, N = 3 SE +/- 0.21, N = 3 56.13 57.55 57.82 -O2 1. (F9X) gfortran options: -O0
Scikit-Learn Benchmark: Covertype Dataset Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Covertype Dataset Benchmark mantic-no-omit-framepointer mantic noble 80 160 240 320 400 SE +/- 3.40, N = 3 SE +/- 4.88, N = 3 SE +/- 2.58, N = 3 370.69 376.15 381.45 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 mantic-no-omit-framepointer mantic 16 32 48 64 80 SE +/- 0.66, N = 3 SE +/- 0.44, N = 3 73.65 71.81 MIN: 68.88 / MAX: 75.03 MIN: 67.31 / MAX: 72.89
PyPerformance Benchmark: pathlib OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pathlib mantic mantic-no-omit-framepointer 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 19.7 20.2
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State mantic-no-omit-framepointer mantic noble 0.3254 0.6508 0.9762 1.3016 1.627 SE +/- 0.001, N = 3 SE +/- 0.004, N = 3 SE +/- 0.006, N = 3 1.411 1.422 1.446
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State mantic mantic-no-omit-framepointer noble 0.3231 0.6462 0.9693 1.2924 1.6155 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 SE +/- 0.003, N = 3 1.402 1.405 1.436
PyPerformance Benchmark: crypto_pyaes OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: crypto_pyaes mantic mantic-no-omit-framepointer 15 30 45 60 75 SE +/- 0.06, N = 3 SE +/- 0.00, N = 3 65.1 66.6
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 mantic mantic-no-omit-framepointer 16 32 48 64 80 SE +/- 0.56, N = 3 SE +/- 0.96, N = 3 73.91 72.27 MIN: 68.9 / MAX: 75.9 MIN: 68.86 / MAX: 76.62
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 mantic-no-omit-framepointer mantic 50 100 150 200 250 SE +/- 1.98, N = 3 SE +/- 0.58, N = 3 205.95 201.41 MIN: 186.96 / MAX: 210.21 MIN: 184.02 / MAX: 203.68
Scikit-Learn Benchmark: Kernel PCA Solvers / Time vs. N Components OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Kernel PCA Solvers / Time vs. N Components noble mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.43, N = 3 SE +/- 0.21, N = 3 SE +/- 0.36, N = 3 37.11 37.24 37.89 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer mantic noble 0.142 0.284 0.426 0.568 0.71 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.006, N = 3 0.618 0.619 0.631
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.15, N = 3 SE +/- 0.30, N = 15 37.36 36.60 MIN: 35.47 / MAX: 37.85 MIN: 33.07 / MAX: 39.53
Scikit-Learn Benchmark: Plot Hierarchical OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot Hierarchical noble mantic-no-omit-framepointer mantic 50 100 150 200 250 SE +/- 2.35, N = 3 SE +/- 0.42, N = 3 SE +/- 0.75, N = 3 207.10 208.39 211.29 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 mantic-no-omit-framepointer mantic 16 32 48 64 80 SE +/- 0.50, N = 3 SE +/- 0.94, N = 3 73.75 72.31 MIN: 68.91 / MAX: 75.15 MIN: 67.38 / MAX: 74.62
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: ResNet-152 mantic-no-omit-framepointer mantic noble 3 6 9 12 15 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 10.00 9.84 9.81 MIN: 8.09 / MAX: 10.27 MIN: 9.6 / MAX: 9.98 MIN: 9.42 / MAX: 9.93
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer mantic noble 0.6003 1.2006 1.8009 2.4012 3.0015 SE +/- 0.006, N = 3 SE +/- 0.006, N = 3 SE +/- 0.006, N = 3 2.620 2.662 2.668
Scikit-Learn Benchmark: Sparsify OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Sparsify noble mantic-no-omit-framepointer mantic 30 60 90 120 150 SE +/- 0.65, N = 3 SE +/- 1.28, N = 5 SE +/- 1.36, N = 5 125.07 125.44 127.28 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.30, N = 9 SE +/- 0.31, N = 15 37.88 37.24 MIN: 35.67 / MAX: 39.63 MIN: 33.97 / MAX: 39.43
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 mantic-no-omit-framepointer mantic 16 32 48 64 80 SE +/- 0.83, N = 3 SE +/- 0.24, N = 3 72.91 71.74 MIN: 68 / MAX: 75.45 MIN: 67.87 / MAX: 72.6
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 mantic-no-omit-framepointer mantic 40 80 120 160 200 SE +/- 2.52, N = 4 SE +/- 1.06, N = 3 202.68 199.46 MIN: 182.69 / MAX: 211.53 MIN: 182.77 / MAX: 206.03
PyPerformance Benchmark: pickle_pure_python OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: pickle_pure_python mantic mantic-no-omit-framepointer 60 120 180 240 300 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 259 263
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 262144 - Benchmark: Isoneutral Mixing mantic mantic-no-omit-framepointer noble 0.0299 0.0598 0.0897 0.1196 0.1495 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.002, N = 3 0.131 0.132 0.133
Scikit-Learn Benchmark: SGDOneClassSVM OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: SGDOneClassSVM mantic mantic-no-omit-framepointer noble 80 160 240 320 400 SE +/- 4.18, N = 3 SE +/- 3.48, N = 7 SE +/- 3.55, N = 3 379.74 382.61 385.38 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.24, N = 3 SE +/- 0.30, N = 15 37.71 37.16 MIN: 35.52 / MAX: 38.25 MIN: 34.12 / MAX: 39.48
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing mantic-no-omit-framepointer noble mantic 0.142 0.284 0.426 0.568 0.71 SE +/- 0.007, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 0.622 0.630 0.631
Scikit-Learn Benchmark: Plot Incremental PCA OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Plot Incremental PCA noble mantic mantic-no-omit-framepointer 7 14 21 28 35 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 30.62 31.01 31.06 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 mantic-no-omit-framepointer noble mantic 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 9.91 9.86 9.77 MIN: 9.19 / MAX: 10.05 MIN: 8.69 / MAX: 9.99 MIN: 9.17 / MAX: 10
Scikit-Learn Benchmark: Feature Expansions OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Feature Expansions mantic mantic-no-omit-framepointer noble 30 60 90 120 150 SE +/- 0.86, N = 3 SE +/- 1.22, N = 3 SE +/- 1.21, N = 3 131.28 133.09 133.14 -O2 1. (F9X) gfortran options: -O0
PyPerformance Benchmark: 2to3 OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: 2to3 mantic mantic-no-omit-framepointer 50 100 150 200 250 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 221 224
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 noble mantic-no-omit-framepointer mantic 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 12.89 12.78 12.72 MIN: 12.36 / MAX: 13.05 MIN: 11.9 / MAX: 12.9 MIN: 11.99 / MAX: 12.8
PyPerformance Benchmark: chaos OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: chaos mantic mantic-no-omit-framepointer 14 28 42 56 70 SE +/- 0.03, N = 3 SE +/- 0.20, N = 3 62.8 63.6
Scikit-Learn Benchmark: Hist Gradient Boosting Threading OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Hist Gradient Boosting Threading mantic mantic-no-omit-framepointer noble 20 40 60 80 100 SE +/- 0.13, N = 3 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 110.22 110.37 111.55 -O2 1. (F9X) gfortran options: -O0
PyPerformance Benchmark: nbody OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: nbody mantic mantic-no-omit-framepointer 20 40 60 80 100 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 76.2 77.1
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State mantic-no-omit-framepointer noble mantic 0.0592 0.1184 0.1776 0.2368 0.296 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.002, N = 3 0.260 0.262 0.263
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 mantic mantic-no-omit-framepointer 16 32 48 64 80 SE +/- 0.96, N = 3 SE +/- 0.74, N = 3 74.15 73.36 MIN: 68.27 / MAX: 75.61 MIN: 68.19 / MAX: 74.63
Numpy Benchmark OpenBenchmarking.org Score, More Is Better Numpy Benchmark noble mantic-no-omit-framepointer mantic 90 180 270 360 450 SE +/- 1.01, N = 3 SE +/- 0.90, N = 3 SE +/- 1.20, N = 3 430.83 428.61 426.28
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 mantic mantic-no-omit-framepointer 16 32 48 64 80 SE +/- 0.96, N = 3 SE +/- 0.20, N = 3 73.01 72.24 MIN: 68.06 / MAX: 75.3 MIN: 68.36 / MAX: 73.14
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 mantic mantic-no-omit-framepointer 40 80 120 160 200 SE +/- 1.69, N = 3 SE +/- 0.33, N = 3 203.18 201.14 MIN: 183.76 / MAX: 207.98 MIN: 183.61 / MAX: 202.73
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: ResNet-50 mantic-no-omit-framepointer mantic noble 6 12 18 24 30 SE +/- 0.16, N = 3 SE +/- 0.10, N = 3 SE +/- 0.06, N = 3 24.35 24.29 24.12 MIN: 23.67 / MAX: 24.87 MIN: 22.24 / MAX: 24.66 MIN: 22.33 / MAX: 24.46
Scikit-Learn Benchmark: 20 Newsgroups / Logistic Regression OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: 20 Newsgroups / Logistic Regression mantic mantic-no-omit-framepointer noble 10 20 30 40 50 SE +/- 0.19, N = 3 SE +/- 0.24, N = 3 SE +/- 0.12, N = 3 41.52 41.73 41.91 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 1.269 2.538 3.807 5.076 6.345 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 5.64 5.63 5.59 MIN: 5.52 / MAX: 5.69 MIN: 5.31 / MAX: 5.68 MIN: 5.46 / MAX: 5.64
PyTorch Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 1.269 2.538 3.807 5.076 6.345 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 5.64 5.63 5.59 MIN: 5.45 / MAX: 5.68 MIN: 5.39 / MAX: 5.71 MIN: 5.31 / MAX: 5.65
PyTorch Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 1.2713 2.5426 3.8139 5.0852 6.3565 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.65 5.61 5.60 MIN: 5.36 / MAX: 5.93 MIN: 5.45 / MAX: 5.66 MIN: 5.37 / MAX: 5.66
PyTorch Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 1.2713 2.5426 3.8139 5.0852 6.3565 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.65 5.62 5.60 MIN: 5.45 / MAX: 5.7 MIN: 5.35 / MAX: 5.66 MIN: 5.32 / MAX: 5.64
PyTorch Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 64 - Model: ResNet-50 mantic-no-omit-framepointer mantic noble 6 12 18 24 30 SE +/- 0.15, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 24.40 24.24 24.19 MIN: 21.6 / MAX: 24.8 MIN: 23.59 / MAX: 24.49 MIN: 22.75 / MAX: 24.73
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State noble mantic-no-omit-framepointer mantic 0.0592 0.1184 0.1776 0.2368 0.296 SE +/- 0.002, N = 3 SE +/- 0.000, N = 3 SE +/- 0.002, N = 3 0.261 0.262 0.263
PyPerformance Benchmark: float OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: float mantic-no-omit-framepointer mantic 15 30 45 60 75 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 66.9 67.4
PyTorch Device: CPU - Batch Size: 512 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 mantic noble mantic-no-omit-framepointer 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 9.87 9.87 9.80 MIN: 9.09 / MAX: 9.96 MIN: 9.21 / MAX: 10 MIN: 9.12 / MAX: 9.98
Scikit-Learn Benchmark: MNIST Dataset OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: MNIST Dataset noble mantic mantic-no-omit-framepointer 15 30 45 60 75 SE +/- 0.67, N = 3 SE +/- 0.82, N = 4 SE +/- 0.47, N = 3 65.42 65.76 65.88 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: CPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 noble mantic-no-omit-framepointer mantic 6 12 18 24 30 SE +/- 0.14, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 24.30 24.28 24.13 MIN: 22.45 / MAX: 24.75 MIN: 22.31 / MAX: 24.53 MIN: 23.58 / MAX: 24.41
Scikit-Learn Benchmark: SAGA OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: SAGA mantic noble mantic-no-omit-framepointer 200 400 600 800 1000 SE +/- 8.69, N = 6 SE +/- 10.35, N = 3 SE +/- 5.60, N = 3 868.02 869.37 873.82 -O2 1. (F9X) gfortran options: -O0
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 mantic-no-omit-framepointer mantic noble 8 16 24 32 40 SE +/- 0.16, N = 3 SE +/- 0.11, N = 3 SE +/- 0.17, N = 3 32.54 32.36 32.34 MIN: 31.64 / MAX: 32.94 MIN: 31.89 / MAX: 32.7 MIN: 28.9 / MAX: 32.83
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: ResNet-50 noble mantic-no-omit-framepointer mantic 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.16, N = 3 SE +/- 0.05, N = 3 24.43 24.38 24.28 MIN: 22.57 / MAX: 24.72 MIN: 22.2 / MAX: 24.87 MIN: 20.22 / MAX: 24.56
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l mantic mantic-no-omit-framepointer 9 18 27 36 45 SE +/- 0.03, N = 3 SE +/- 0.33, N = 8 37.43 37.22 MIN: 35.81 / MAX: 38.02 MIN: 34.99 / MAX: 39.08
PyTorch Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 1.269 2.538 3.807 5.076 6.345 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 5.64 5.61 5.61 MIN: 5.29 / MAX: 5.68 MIN: 5.44 / MAX: 5.65 MIN: 5.46 / MAX: 5.67
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 16 - Model: ResNet-152 mantic-no-omit-framepointer mantic noble 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 9.93 9.88 9.88 MIN: 9.39 / MAX: 10.01 MIN: 9.31 / MAX: 10.01 MIN: 9.15 / MAX: 9.98
PyTorch Device: CPU - Batch Size: 64 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 64 - Model: ResNet-152 mantic-no-omit-framepointer mantic noble 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 9.91 9.88 9.87 MIN: 8.69 / MAX: 10.08 MIN: 8.8 / MAX: 9.98 MIN: 8.61 / MAX: 9.96
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 mantic mantic-no-omit-framepointer noble 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.06, N = 3 24.42 24.37 24.33 MIN: 20.15 / MAX: 24.74 MIN: 23.76 / MAX: 24.81 MIN: 22.79 / MAX: 24.66
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 mantic-no-omit-framepointer mantic 50 100 150 200 250 SE +/- 1.46, N = 15 SE +/- 2.67, N = 3 211.46 210.88 MIN: 192.13 / MAX: 223.01 MIN: 195.21 / MAX: 218.16
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 mantic-no-omit-framepointer mantic 40 80 120 160 200 SE +/- 1.21, N = 3 SE +/- 1.76, N = 3 203.22 202.72 MIN: 185.88 / MAX: 206.71 MIN: 183.1 / MAX: 207.93
PyTorch Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l mantic-no-omit-framepointer mantic noble 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.32 7.31 7.31 MIN: 7.23 / MAX: 7.38 MIN: 7.16 / MAX: 7.34 MIN: 7.07 / MAX: 7.36
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 mantic mantic-no-omit-framepointer 40 80 120 160 200 SE +/- 0.25, N = 3 SE +/- 0.96, N = 3 200.30 200.17 MIN: 182.88 / MAX: 202.36 MIN: 183.43 / MAX: 203.55
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 65536 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 65536 - Benchmark: Equation of State mantic mantic-no-omit-framepointer noble 0.0034 0.0068 0.0102 0.0136 0.017 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 7 0.015 0.015 0.015
PyHPC Benchmarks Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 16384 - Benchmark: Equation of State mantic mantic-no-omit-framepointer noble 0.0007 0.0014 0.0021 0.0028 0.0035 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.003 0.003 0.003
Scikit-Learn Benchmark: Isolation Forest OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 1.2.2 Benchmark: Isolation Forest mantic noble mantic-no-omit-framepointer 70 140 210 280 350 SE +/- 1.30, N = 3 SE +/- 2.83, N = 3 SE +/- 51.04, N = 9 289.37 314.03 336.37 -O2 1. (F9X) gfortran options: -O0
PyHPC Benchmarks Device: GPU - Backend: Numpy - Project Size: 16384 - Benchmark: Equation of State OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: GPU - Backend: Numpy - Project Size: 16384 - Benchmark: Equation of State mantic-no-omit-framepointer mantic noble 0.0007 0.0014 0.0021 0.0028 0.0035 SE +/- 0.000, N = 15 SE +/- 0.000, N = 3 SE +/- 0.000, N = 12 0.002 0.003 0.003
Phoronix Test Suite v10.8.5