pytorch-test AMD Ryzen 7 5700X 8-Core testing with a MSI PRO B550M-P GEN3 (MS-7D95) v1.0 (1.60 BIOS) and NVIDIA GeForce RTX 4070 SUPER 12GB on Debian via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2403306-NE-PYTORCHTE32 .
pytorch-test Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Compiler File-System Screen Resolution All options AMD Ryzen 7 5700X 8-Core @ 3.40GHz (8 Cores / 16 Threads) MSI PRO B550M-P GEN3 (MS-7D95) v1.0 (1.60 BIOS) AMD Starship/Matisse 128GB 256GB ARDOR GAMING m.2 NVME 256Gb AL1282 + 4001GB Seagate ST4000VX016-3CV1 + 512GB Apacer AS350 512 NVIDIA GeForce RTX 4070 SUPER 12GB NVIDIA Device 22bc PHL 242V8 Realtek RTL8111/8168/8211/8411 Debian 6.6.15-amd64 (x86_64) Xfce 4.18 X Server 1.21.1.11 NVIDIA 550.54.15 4.6.0 OpenCL 3.0 CUDA 12.4.89 + OpenCL 3.0 PoCL 5.0+debian Linux +Asserts RELOC SPIR LLVM 16.0.6 SLEEF DISTRO POCL_DEBUG GCC 13.2.0 + CUDA 12.0 ext4 4480x1440 OpenBenchmarking.org - Transparent Huge Pages: always - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa20120e - Python 3.11.8 - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
pytorch-test pytorch: CPU - 1 - ResNet-50 pytorch: CPU - 1 - ResNet-152 pytorch: CPU - 16 - ResNet-50 pytorch: CPU - 32 - ResNet-50 pytorch: CPU - 64 - ResNet-50 pytorch: CPU - 16 - ResNet-152 pytorch: CPU - 256 - ResNet-50 pytorch: CPU - 32 - ResNet-152 pytorch: CPU - 512 - ResNet-50 pytorch: CPU - 64 - ResNet-152 pytorch: CPU - 256 - ResNet-152 pytorch: CPU - 512 - ResNet-152 pytorch: CPU - 1 - Efficientnet_v2_l pytorch: CPU - 16 - Efficientnet_v2_l pytorch: CPU - 32 - Efficientnet_v2_l pytorch: CPU - 64 - Efficientnet_v2_l pytorch: CPU - 256 - Efficientnet_v2_l pytorch: CPU - 512 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 1 - ResNet-50 pytorch: NVIDIA CUDA GPU - 1 - ResNet-152 pytorch: NVIDIA CUDA GPU - 16 - ResNet-50 pytorch: NVIDIA CUDA GPU - 32 - ResNet-50 pytorch: NVIDIA CUDA GPU - 64 - ResNet-50 pytorch: NVIDIA CUDA GPU - 16 - ResNet-152 pytorch: NVIDIA CUDA GPU - 256 - ResNet-50 pytorch: NVIDIA CUDA GPU - 32 - ResNet-152 pytorch: NVIDIA CUDA GPU - 512 - ResNet-50 pytorch: NVIDIA CUDA GPU - 64 - ResNet-152 pytorch: NVIDIA CUDA GPU - 256 - ResNet-152 pytorch: NVIDIA CUDA GPU - 512 - ResNet-152 pytorch: NVIDIA CUDA GPU - 1 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 16 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 32 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 64 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 256 - Efficientnet_v2_l pytorch: NVIDIA CUDA GPU - 512 - Efficientnet_v2_l All options 44.60 18.16 28.47 28.33 27.77 11.47 27.81 11.52 27.94 11.50 11.41 11.32 10.98 7.74 7.61 7.60 7.50 7.44 212.31 76.01 214.58 217.50 217.22 73.74 212.89 76.28 215.10 75.03 76.76 77.12 40.73 38.81 39.03 39.00 39.65 37.97 OpenBenchmarking.org
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 All options 10 20 30 40 50 SE +/- 0.50, N = 3 44.60 MIN: 31.6 / MAX: 46.25
PyTorch Device: CPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 All options 4 8 12 16 20 SE +/- 0.12, N = 3 18.16 MIN: 14.49 / MAX: 18.57
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: ResNet-50 All options 7 14 21 28 35 SE +/- 0.32, N = 5 28.47 MIN: 19.33 / MAX: 30.1
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: ResNet-50 All options 7 14 21 28 35 SE +/- 0.29, N = 3 28.33 MIN: 19.28 / MAX: 29.17
PyTorch Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: ResNet-50 All options 7 14 21 28 35 SE +/- 0.21, N = 10 27.77 MIN: 16.01 / MAX: 29.49
PyTorch Device: CPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: ResNet-152 All options 3 6 9 12 15 SE +/- 0.07, N = 3 11.47 MIN: 8.19 / MAX: 11.89
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 All options 7 14 21 28 35 SE +/- 0.22, N = 15 27.81 MIN: 12.38 / MAX: 30.18
PyTorch Device: CPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: ResNet-152 All options 3 6 9 12 15 SE +/- 0.02, N = 3 11.52 MIN: 9.36 / MAX: 11.86
PyTorch Device: CPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 All options 7 14 21 28 35 SE +/- 0.32, N = 15 27.94 MIN: 16.81 / MAX: 29.62
PyTorch Device: CPU - Batch Size: 64 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: ResNet-152 All options 3 6 9 12 15 SE +/- 0.10, N = 3 11.50 MIN: 8.81 / MAX: 11.87
PyTorch Device: CPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: ResNet-152 All options 3 6 9 12 15 SE +/- 0.02, N = 3 11.41 MIN: 8.11 / MAX: 11.89
PyTorch Device: CPU - Batch Size: 512 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 All options 3 6 9 12 15 SE +/- 0.10, N = 3 11.32 MIN: 7.82 / MAX: 12.05
PyTorch Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l All options 3 6 9 12 15 SE +/- 0.04, N = 3 10.98 MIN: 8.22 / MAX: 11.42
PyTorch Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l All options 2 4 6 8 10 SE +/- 0.01, N = 3 7.74 MIN: 5.95 / MAX: 7.85
PyTorch Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l All options 2 4 6 8 10 SE +/- 0.10, N = 3 7.61 MIN: 5.99 / MAX: 7.8
PyTorch Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l All options 2 4 6 8 10 SE +/- 0.05, N = 3 7.60 MIN: 5.08 / MAX: 7.86
PyTorch Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l All options 2 4 6 8 10 SE +/- 0.09, N = 3 7.50 MIN: 5.5 / MAX: 7.79
PyTorch Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l All options 2 4 6 8 10 SE +/- 0.07, N = 9 7.44 MIN: 4.46 / MAX: 7.85
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 2.38, N = 3 212.31 MIN: 110.61 / MAX: 225.39
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152 All options 20 40 60 80 100 SE +/- 0.73, N = 3 76.01 MIN: 44.66 / MAX: 78.91
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 2.06, N = 3 214.58 MIN: 172.29 / MAX: 221
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 0.92, N = 3 217.50 MIN: 172.82 / MAX: 221.5
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 0.25, N = 3 217.22 MIN: 199.94 / MAX: 220.82
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152 All options 16 32 48 64 80 SE +/- 0.08, N = 3 73.74 MIN: 43.03 / MAX: 79.24
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 2.38, N = 4 212.89 MIN: 127.22 / MAX: 220.8
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152 All options 20 40 60 80 100 SE +/- 0.91, N = 3 76.28 MIN: 39.08 / MAX: 78.37
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50 All options 50 100 150 200 250 SE +/- 2.52, N = 3 215.10 MIN: 122.92 / MAX: 222.21
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152 All options 20 40 60 80 100 SE +/- 0.79, N = 5 75.03 MIN: 50.37 / MAX: 78.56
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152 All options 20 40 60 80 100 SE +/- 0.39, N = 3 76.76 MIN: 58.4 / MAX: 78.75
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152 All options 20 40 60 80 100 SE +/- 0.08, N = 3 77.12 MIN: 69.63 / MAX: 78.12
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.08, N = 3 40.73 MIN: 37.27 / MAX: 41.27
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.12, N = 3 38.81 MIN: 33.4 / MAX: 40.1
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.56, N = 3 39.03 MIN: 35.91 / MAX: 40.16
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.29, N = 3 39.00 MIN: 36.15 / MAX: 40
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.05, N = 3 39.65 MIN: 36.18 / MAX: 40.35
PyTorch Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.2.1 Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l All options 9 18 27 36 45 SE +/- 0.41, N = 5 37.97 MIN: 24.02 / MAX: 39.94
Phoronix Test Suite v10.8.4