Apple M1 Max testing with a Apple MacBook Pro and Apple M1 Max on macOS 12.1 via the Phoronix Test Suite.
sys76-kudu-ML: AMD Ryzen 9 5900HX testing with a System76 Kudu (1.07.09RSA1 BIOS) and AMD Cezanne on Pop 21.10 via the Phoronix Test Suite.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2202161-NE-MBPM1MAXM40
MBP M1 Max Machine Learning, sys76-kudu-ML,
"NCNN 20210720 - Target: Vulkan GPU - Model: alexnet",
Lower Results Are Better
"MBP M1 Max Machine Learning",29.89,29.88,29.89
"ML Tests",6.27,6.34,6.35
"Mobile Neural Network 1.2 - Model: MobileNetV2_224",
Lower Results Are Better
"MBP M1 Max Machine Learning",10.331,9.469,10.474,10.425,10.977,11.069,11.167,11.026,11.151
"ML Tests",2.364,2.423,2.374
"NCNN 20210720 - Target: Vulkan GPU - Model: resnet50",
Lower Results Are Better
"MBP M1 Max Machine Learning",43.09,43.09,43.05
"ML Tests",13.02,13.13,13.21
"NCNN 20210720 - Target: Vulkan GPU - Model: googlenet",
Lower Results Are Better
"MBP M1 Max Machine Learning",24.9,24.9,24.9
"ML Tests",8.25,8.84,9.09
"NCNN 20210720 - Target: Vulkan GPU - Model: resnet18",
Lower Results Are Better
"MBP M1 Max Machine Learning",16.82,16.78,16.8
"ML Tests",6.17,6.17,5.94
"NCNN 20210720 - Target: CPU - Model: alexnet",
Lower Results Are Better
"MBP M1 Max Machine Learning",30.04,29.88,29.88
"ML Tests",14.52,14.46,14.66
"NCNN 20210720 - Target: Vulkan GPU - Model: mobilenet",
Lower Results Are Better
"MBP M1 Max Machine Learning",20.33,20.29,20.28
"ML Tests",10.26,10.13,10.43
"NCNN 20210720 - Target: CPU - Model: googlenet",
Lower Results Are Better
"MBP M1 Max Machine Learning",25.11,24.89,24.89
"ML Tests",14.11,13.18,13.93
"NCNN 20210720 - Target: CPU - Model: resnet50",
Lower Results Are Better
"MBP M1 Max Machine Learning",43.29,43.09,43.1
"ML Tests",25.06,25.28,25.16
"NCNN 20210720 - Target: CPU - Model: efficientnet-b0",
Lower Results Are Better
"MBP M1 Max Machine Learning",8.78,8.67,8.63
"ML Tests",5.21,5.21,5.23
"NCNN 20210720 - Target: CPU - Model: mnasnet",
Lower Results Are Better
"MBP M1 Max Machine Learning",5.46,5.37,5.37
"ML Tests",3.31,3.23,3.2
"NCNN 20210720 - Target: Vulkan GPU - Model: yolov4-tiny",
Lower Results Are Better
"MBP M1 Max Machine Learning",30.36,30.43,30.2
"ML Tests",18.99,19.46,18.02
"NCNN 20210720 - Target: Vulkan GPU - Model: vgg16",
Lower Results Are Better
"MBP M1 Max Machine Learning",70.93,70.85,70.88
"ML Tests",43.94,43.9,44.14
"NCNN 20210720 - Target: Vulkan GPU - Model: mnasnet",
Lower Results Are Better
"MBP M1 Max Machine Learning",5.37,5.37,5.37
"ML Tests",3.78,3.99
"NCNN 20210720 - Target: CPU - Model: blazeface",
Lower Results Are Better
"MBP M1 Max Machine Learning",1.67,1.64,1.64
"ML Tests",1.2,1.18,1.21
"NCNN 20210720 - Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2",
Lower Results Are Better
"MBP M1 Max Machine Learning",5.31,5.29,5.29
"ML Tests",3.76,3.95,3.94
"NCNN 20210720 - Target: Vulkan GPU - Model: regnety_400m",
Lower Results Are Better
"MBP M1 Max Machine Learning",7.18,7.19,7.19
"ML Tests",5.21,5.24,5.39
"NCNN 20210720 - Target: Vulkan GPU - Model: squeezenet_ssd",
Lower Results Are Better
"MBP M1 Max Machine Learning",20.47,20.63,20.56
"ML Tests",14.88,15.13,16.06
"NCNN 20210720 - Target: CPU-v2-v2 - Model: mobilenet-v2",
Lower Results Are Better
"MBP M1 Max Machine Learning",5.39,5.3,5.3
"ML Tests",3.98,4.02,3.97
"NCNN 20210720 - Target: CPU-v3-v3 - Model: mobilenet-v3",
Lower Results Are Better
"MBP M1 Max Machine Learning",4.42,4.34,4.33
"ML Tests",3.36,3.42,3.44
"NCNN 20210720 - Target: CPU - Model: mobilenet",
Lower Results Are Better
"MBP M1 Max Machine Learning",20.36,20.29,20.3
"ML Tests",15.85,16.13,15.87
"NCNN 20210720 - Target: CPU - Model: shufflenet-v2",
Lower Results Are Better
"MBP M1 Max Machine Learning",3.51,3.45,3.45
"ML Tests",2.68,2.77,2.81
"NCNN 20210720 - Target: Vulkan GPU - Model: blazeface",
Lower Results Are Better
"MBP M1 Max Machine Learning",1.64,1.64,1.64
"ML Tests",1.34,1.38,1.34
"NCNN 20210720 - Target: CPU - Model: yolov4-tiny",
Lower Results Are Better
"MBP M1 Max Machine Learning",30.2,30.23,30.29
"ML Tests",24.78,25.18,24.96
"NCNN 20210720 - Target: Vulkan GPU - Model: efficientnet-b0",
Lower Results Are Better
"MBP M1 Max Machine Learning",8.71,8.74,8.67
"ML Tests",10.14,9.94,10.12
"NCNN 20210720 - Target: Vulkan GPU - Model: shufflenet-v2",
Lower Results Are Better
"MBP M1 Max Machine Learning",3.48,3.45,3.46
"ML Tests",2.95,2.95,3.16
"NCNN 20210720 - Target: CPU - Model: squeezenet_ssd",
Lower Results Are Better
"MBP M1 Max Machine Learning",20.55,20.44,20.6
"ML Tests",18.27,18.74,18.68
"NCNN 20210720 - Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3",
Lower Results Are Better
"MBP M1 Max Machine Learning",4.35,4.35,4.34
"ML Tests",4.92,4.5,4.66
"NCNN 20210720 - Target: CPU - Model: resnet18",
Lower Results Are Better
"MBP M1 Max Machine Learning",16.89,16.78,16.78
"ML Tests",16.22,14.92,16.2
"NCNN 20210720 - Target: CPU - Model: regnety_400m",
Lower Results Are Better
"MBP M1 Max Machine Learning",7.18,7.18,7.19
"ML Tests",6.95,6.81,6.93
"NCNN 20210720 - Target: CPU - Model: vgg16",
Lower Results Are Better
"MBP M1 Max Machine Learning",71.3,70.9,70.82
"ML Tests",71.83,72.28,71.8
"Mlpack Benchmark - Benchmark: scikit_linearridgeregression",
Lower Results Are Better
"ML Tests",2.0961380004883,2.0926539897919,2.1222839355469
"Mlpack Benchmark - Benchmark: scikit_svm",
Lower Results Are Better
"ML Tests",17.610609531403,17.621352910995,17.570944547653
"Mlpack Benchmark - Benchmark: scikit_qda",
Lower Results Are Better
"ML Tests",65.696218013763,65.6326816082,65.738785982132
"Mlpack Benchmark - Benchmark: scikit_ica",
Lower Results Are Better
"ML Tests",48.625242710114,48.206023693085,48.363018989563
"ECP-CANDLE 0.4 - Benchmark: P3B2",
Lower Results Are Better
"ML Tests",
"ECP-CANDLE 0.4 - Benchmark: P3B1",
Lower Results Are Better
"ML Tests",
"ECP-CANDLE 0.4 - Benchmark: P1B2",
Lower Results Are Better
"ML Tests",
"PlaidML - FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU",
Higher Results Are Better
"ML Tests",6.92,6.85,6.88
"PlaidML - FP16: No - Mode: Inference - Network: VGG16 - Device: CPU",
Higher Results Are Better
"ML Tests",12.36,12.43,12.61
"TNN 0.3 - Target: CPU - Model: SqueezeNet v1.1",
Lower Results Are Better
"ML Tests",222.245,222.151,222.583
"TNN 0.3 - Target: CPU - Model: SqueezeNet v2",
Lower Results Are Better
"ML Tests",54.401,55.368,56.533
"TNN 0.3 - Target: CPU - Model: MobileNet v2",
Lower Results Are Better
"ML Tests",249.92,248.682,249.829
"TNN 0.3 - Target: CPU - Model: DenseNet",
Lower Results Are Better
"ML Tests",2734.957,2735.798,2737.764
"Caffe 2020-02-13 - Model: GoogleNet - Acceleration: CPU - Iterations: 1000",
Lower Results Are Better
"ML Tests",867869,868934,869471
"Caffe 2020-02-13 - Model: GoogleNet - Acceleration: CPU - Iterations: 200",
Lower Results Are Better
"ML Tests",173960,174018,173036
"Caffe 2020-02-13 - Model: GoogleNet - Acceleration: CPU - Iterations: 100",
Lower Results Are Better
"ML Tests",86747,86566,86389
"Caffe 2020-02-13 - Model: AlexNet - Acceleration: CPU - Iterations: 1000",
Lower Results Are Better
"ML Tests",325390,325438,326823
"Caffe 2020-02-13 - Model: AlexNet - Acceleration: CPU - Iterations: 200",
Lower Results Are Better
"ML Tests",65911,65740,66306
"Caffe 2020-02-13 - Model: AlexNet - Acceleration: CPU - Iterations: 100",
Lower Results Are Better
"ML Tests",33446,33473,33569
"TensorFlow Lite 2020-08-23 - Model: Inception ResNet V2",
Lower Results Are Better
"ML Tests",2476730,2479930,2480580
"TensorFlow Lite 2020-08-23 - Model: Mobilenet Quant",
Lower Results Are Better
"ML Tests",141105,141237,141181
"TensorFlow Lite 2020-08-23 - Model: Mobilenet Float",
Lower Results Are Better
"ML Tests",127626,128167,127661
"TensorFlow Lite 2020-08-23 - Model: NASNet Mobile",
Lower Results Are Better
"ML Tests",152578,152481,151498
"TensorFlow Lite 2020-08-23 - Model: Inception V4",
Lower Results Are Better
"ML Tests",2753040,2748260,2747570
"TensorFlow Lite 2020-08-23 - Model: SqueezeNet",
Lower Results Are Better
"ML Tests",189572,189772,189949
"RNNoise 2020-06-28 - ",
Lower Results Are Better
"ML Tests",16.167,16.135,16.11
"R Benchmark - ",
Lower Results Are Better
"ML Tests",0.12921074116417,0.12882049077911,0.1298200310973
"LeelaChessZero 0.28 - Backend: BLAS",
Higher Results Are Better
"ML Tests",579,544,567,580,551,566,557
"DeepSpeech 0.6 - Acceleration: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",74.11764,74.49802,74.70562
"Numpy Benchmark - ",
Higher Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",423.24,423.35,420.77
"oneDNN 2.1.2 - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2.96524,2.99699,2.99755
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2244.25,2422.62,2214.81,2223.58,2225.18,2237.74,2225.5,2219.74,2201.09,2207.76,2244.39,2232.2,2218.09,2210.17
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",3585.28,3577.51,3568.22
"oneDNN 2.1.2 - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",4.5834,4.59494,4.60196
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2221.83,2221.84,2240.84
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",3581.17,3595.17,3585.16
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2216.92,2220.92,2219.55
"oneDNN 2.1.2 - Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",3581.53,3590.34,3565.12
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",3.29853,3.20668,3.23831
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2.1086,2.12315,2.12139
"oneDNN 2.1.2 - Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",23.7862,23.722,23.7939
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",6.72661,6.74949,6.76066
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",8.2983,8.39678,8.34858
"oneDNN 2.1.2 - Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",22.7585,22.8593,22.7599
"oneDNN 2.1.2 - Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",2.69651,2.68944,2.69036
"oneDNN 2.1.2 - Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",1.61167,1.64133,1.63656
"oneDNN 2.1.2 - Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",12.0592,12.0775,12.141
"oneDNN 2.1.2 - Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",4.1774,4.42211,4.35687,4.14753,4.2045,4.22199,4.27944
"OpenCV 4.5.4 - Test: DNN - Deep Neural Network",
Lower Results Are Better
"ML Tests",12490,12523,14957,15100,13785,13644,13681,14435,15056,14259,14629,13587,13871,11414,13371
"ONNX Runtime 1.10 - Model: super-resolution-10 - Device: CPU",
Higher Results Are Better
"ML Tests",
"ONNX Runtime 1.10 - Model: shufflenet-v2-10 - Device: CPU",
Higher Results Are Better
"ML Tests",
"ONNX Runtime 1.10 - Model: fcn-resnet101-11 - Device: CPU",
Higher Results Are Better
"ML Tests",
"ONNX Runtime 1.10 - Model: yolov4 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Age Gender Recognition Retail 0013 FP32 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Age Gender Recognition Retail 0013 FP16 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Person Detection 0106 FP32 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Person Detection 0106 FP16 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Face Detection 0106 FP32 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Face Detection 0106 FP16 - Device: Intel GPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Person Detection 0106 FP32 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Person Detection 0106 FP16 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Face Detection 0106 FP32 - Device: CPU",
Higher Results Are Better
"ML Tests",
"OpenVINO 2021.1 - Model: Face Detection 0106 FP16 - Device: CPU",
Higher Results Are Better
"ML Tests",
"Tensorflow - Build: Cifar10",
Lower Results Are Better
"ML Tests",
"AI Benchmark Alpha 0.1.2 - ",
Higher Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Numenta Anomaly Benchmark 1.1 - Detector: Bayesian Changepoint",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Numenta Anomaly Benchmark 1.1 - Detector: Earthgecko Skyline",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Numenta Anomaly Benchmark 1.1 - Detector: Windowed Gaussian",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Numenta Anomaly Benchmark 1.1 - Detector: Relative Entropy",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Numenta Anomaly Benchmark 1.1 - Detector: EXPoSE",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"Mobile Neural Network 1.2 - Model: inception-v3",
Lower Results Are Better
"MBP M1 Max Machine Learning",90.597,64.622,60.799,45.224,44.597,44.546,44.596,44.792,84.503
"ML Tests",32.394,31.321,31.014
"Mobile Neural Network 1.2 - Model: mobilenet-v1-1.0",
Lower Results Are Better
"MBP M1 Max Machine Learning",10.741,8.541,9.358,7.474,7.521,7.691,7.468,7.473,7.576
"ML Tests",2.405,2.445,2.471
"Mobile Neural Network 1.2 - Model: SqueezeNetV1.0",
Lower Results Are Better
"MBP M1 Max Machine Learning",15.04,10.525,10.054,9.069,9.001,9.322,9.131,8.606,8.955
"ML Tests",4.623,4.526,4.49
"Mobile Neural Network 1.2 - Model: resnet-v2-50",
Lower Results Are Better
"MBP M1 Max Machine Learning",60.042,59.601,57.493,36.453,33.594,33.74,33.531,33.456,33.94
"ML Tests",22.604,22.407,22.311
"Mobile Neural Network 1.2 - Model: squeezenetv1.1",
Lower Results Are Better
"MBP M1 Max Machine Learning",9.877,6.458,6.374,7.026,6.895,7.218,7.259,6.967,7.39
"ML Tests",2.786,2.811,2.813
"Mobile Neural Network 1.2 - Model: mobilenetV3",
Lower Results Are Better
"MBP M1 Max Machine Learning",7.544,7.2,6.907,10.149,10.125,10.219,10.084,9.989,10.151
"ML Tests",1.209,1.193,1.204
"oneDNN 2.1.2 - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"oneDNN 2.1.2 - Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"oneDNN 2.1.2 - Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"oneDNN 2.1.2 - Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",
"oneDNN 2.1.2 - Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU",
Lower Results Are Better
"MBP M1 Max Machine Learning",
"ML Tests",