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
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.
ML Tests:
Processor: AMD Ryzen 9 5900HX @ 3.30GHz (8 Cores / 16 Threads), Motherboard: System76 Kudu (1.07.09RSA1 BIOS), Chipset: AMD Renoir/Cezanne, Memory: 16GB, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Cezanne (2100/400MHz), Audio: AMD Renoir Radeon HD Audio, Network: Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX200
OS: Pop 21.10, Kernel: 5.15.15-76051515-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server 1.20.13, OpenGL: 4.6 Mesa 21.2.2 (LLVM 12.0.1), Vulkan: 1.2.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
MBP M1 Max Machine Learning:
Processor: Apple M1 Max (10 Cores), Motherboard: Apple MacBook Pro, Memory: 64GB, Disk: 1859GB, Graphics: Apple M1 Max, Monitor: Color LCD
OS: macOS 12.1, Kernel: 21.2.0 (arm64), OpenCL: OpenCL 1.2 (Nov 13 2021 00:45:09), Compiler: GCC 13.0.0 + Clang 13.0.0, File-System: APFS, Screen Resolution: 3456x2234
Caffe 2020-02-13
Model: GoogleNet - Acceleration: CPU - Iterations: 1000
Milli-Seconds < Lower Is Better
ML Tests . 868758 |============================================================
LeelaChessZero 0.28
Backend: BLAS
Nodes Per Second > Higher Is Better
ML Tests . 563 |===============================================================
ECP-CANDLE 0.4
Benchmark: P3B1
Seconds < Lower Is Better
ML Tests . 1463.72 |===========================================================
Mobile Neural Network 1.2
Model: inception-v3
ms < Lower Is Better
ML Tests .................... 31.58 |=======================
MBP M1 Max Machine Learning . 58.25 |==========================================
Mobile Neural Network 1.2
Model: mobilenet-v1-1.0
ms < Lower Is Better
ML Tests .................... 2.440 |============
MBP M1 Max Machine Learning . 8.205 |==========================================
Mobile Neural Network 1.2
Model: MobileNetV2_224
ms < Lower Is Better
ML Tests .................... 2.387 |=========
MBP M1 Max Machine Learning . 10.677 |=========================================
Mobile Neural Network 1.2
Model: SqueezeNetV1.0
ms < Lower Is Better
ML Tests .................... 4.546 |===================
MBP M1 Max Machine Learning . 9.967 |==========================================
Mobile Neural Network 1.2
Model: resnet-v2-50
ms < Lower Is Better
ML Tests .................... 22.44 |======================
MBP M1 Max Machine Learning . 42.43 |==========================================
Mobile Neural Network 1.2
Model: squeezenetv1.1
ms < Lower Is Better
ML Tests .................... 2.803 |================
MBP M1 Max Machine Learning . 7.274 |==========================================
Mobile Neural Network 1.2
Model: mobilenetV3
ms < Lower Is Better
ML Tests .................... 1.202 |======
MBP M1 Max Machine Learning . 9.152 |==========================================
Caffe 2020-02-13
Model: AlexNet - Acceleration: CPU - Iterations: 1000
Milli-Seconds < Lower Is Better
ML Tests . 325884 |============================================================
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
FPS > Higher Is Better
ML Tests . 6.88 |==============================================================
ECP-CANDLE 0.4
Benchmark: P3B2
Seconds < Lower Is Better
ML Tests . 730.74 |============================================================
PlaidML
FP16: No - Mode: Inference - Network: VGG16 - Device: CPU
FPS > Higher Is Better
ML Tests . 12.47 |=============================================================
TNN 0.3
Target: CPU - Model: DenseNet
ms < Lower Is Better
ML Tests . 2736.17 |===========================================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2237.65 |========================================
Caffe 2020-02-13
Model: GoogleNet - Acceleration: CPU - Iterations: 200
Milli-Seconds < Lower Is Better
ML Tests . 173671 |============================================================
TensorFlow Lite 2020-08-23
Model: Inception V4
Microseconds < Lower Is Better
ML Tests . 2749623 |===========================================================
TensorFlow Lite 2020-08-23
Model: Inception ResNet V2
Microseconds < Lower Is Better
ML Tests . 2479080 |===========================================================
Mlpack Benchmark
Benchmark: scikit_qda
Seconds < Lower Is Better
ML Tests . 65.69 |=============================================================
Numpy Benchmark
Score > Higher Is Better
ML Tests .................... 422.45 |=========================================
NCNN 20210720
Target: CPU - Model: regnety_400m
ms < Lower Is Better
ML Tests .................... 6.90 |=========================================
MBP M1 Max Machine Learning . 7.18 |===========================================
NCNN 20210720
Target: CPU - Model: squeezenet_ssd
ms < Lower Is Better
ML Tests .................... 18.56 |======================================
MBP M1 Max Machine Learning . 20.53 |==========================================
NCNN 20210720
Target: CPU - Model: yolov4-tiny
ms < Lower Is Better
ML Tests .................... 24.97 |===================================
MBP M1 Max Machine Learning . 30.24 |==========================================
NCNN 20210720
Target: CPU - Model: resnet50
ms < Lower Is Better
ML Tests .................... 25.17 |========================
MBP M1 Max Machine Learning . 43.16 |==========================================
NCNN 20210720
Target: CPU - Model: alexnet
ms < Lower Is Better
ML Tests .................... 14.55 |====================
MBP M1 Max Machine Learning . 29.93 |==========================================
NCNN 20210720
Target: CPU - Model: resnet18
ms < Lower Is Better
ML Tests .................... 15.78 |=======================================
MBP M1 Max Machine Learning . 16.82 |==========================================
NCNN 20210720
Target: CPU - Model: vgg16
ms < Lower Is Better
ML Tests .................... 71.97 |==========================================
MBP M1 Max Machine Learning . 71.01 |=========================================
NCNN 20210720
Target: CPU - Model: googlenet
ms < Lower Is Better
ML Tests .................... 13.74 |=======================
MBP M1 Max Machine Learning . 24.96 |==========================================
NCNN 20210720
Target: CPU - Model: blazeface
ms < Lower Is Better
ML Tests .................... 1.20 |===============================
MBP M1 Max Machine Learning . 1.65 |===========================================
NCNN 20210720
Target: CPU - Model: efficientnet-b0
ms < Lower Is Better
ML Tests .................... 5.22 |==========================
MBP M1 Max Machine Learning . 8.69 |===========================================
NCNN 20210720
Target: CPU - Model: mnasnet
ms < Lower Is Better
ML Tests .................... 3.25 |==========================
MBP M1 Max Machine Learning . 5.40 |===========================================
NCNN 20210720
Target: CPU - Model: shufflenet-v2
ms < Lower Is Better
ML Tests .................... 2.75 |==================================
MBP M1 Max Machine Learning . 3.47 |===========================================
NCNN 20210720
Target: CPU-v3-v3 - Model: mobilenet-v3
ms < Lower Is Better
ML Tests .................... 3.41 |==================================
MBP M1 Max Machine Learning . 4.36 |===========================================
NCNN 20210720
Target: CPU-v2-v2 - Model: mobilenet-v2
ms < Lower Is Better
ML Tests .................... 3.99 |================================
MBP M1 Max Machine Learning . 5.33 |===========================================
NCNN 20210720
Target: CPU - Model: mobilenet
ms < Lower Is Better
ML Tests .................... 15.95 |=================================
MBP M1 Max Machine Learning . 20.32 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: regnety_400m
ms < Lower Is Better
ML Tests .................... 5.28 |================================
MBP M1 Max Machine Learning . 7.19 |===========================================
NCNN 20210720
Target: Vulkan GPU - Model: squeezenet_ssd
ms < Lower Is Better
ML Tests .................... 15.36 |===============================
MBP M1 Max Machine Learning . 20.55 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: yolov4-tiny
ms < Lower Is Better
ML Tests .................... 18.82 |==========================
MBP M1 Max Machine Learning . 30.33 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: resnet50
ms < Lower Is Better
ML Tests .................... 13.12 |=============
MBP M1 Max Machine Learning . 43.08 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: alexnet
ms < Lower Is Better
ML Tests .................... 6.32 |=========
MBP M1 Max Machine Learning . 29.89 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: resnet18
ms < Lower Is Better
ML Tests .................... 6.09 |===============
MBP M1 Max Machine Learning . 16.80 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: vgg16
ms < Lower Is Better
ML Tests .................... 43.99 |==========================
MBP M1 Max Machine Learning . 70.89 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: googlenet
ms < Lower Is Better
ML Tests .................... 8.73 |===============
MBP M1 Max Machine Learning . 24.90 |==========================================
NCNN 20210720
Target: Vulkan GPU - Model: blazeface
ms < Lower Is Better
ML Tests .................... 1.35 |===================================
MBP M1 Max Machine Learning . 1.64 |===========================================
NCNN 20210720
Target: Vulkan GPU - Model: efficientnet-b0
ms < Lower Is Better
ML Tests .................... 10.07 |==========================================
MBP M1 Max Machine Learning . 8.71 |====================================
NCNN 20210720
Target: Vulkan GPU - Model: mnasnet
ms < Lower Is Better
ML Tests .................... 3.89 |===============================
MBP M1 Max Machine Learning . 5.37 |===========================================
NCNN 20210720
Target: Vulkan GPU - Model: shufflenet-v2
ms < Lower Is Better
ML Tests .................... 3.02 |======================================
MBP M1 Max Machine Learning . 3.46 |===========================================
NCNN 20210720
Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3
ms < Lower Is Better
ML Tests .................... 4.69 |===========================================
MBP M1 Max Machine Learning . 4.35 |========================================
NCNN 20210720
Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2
ms < Lower Is Better
ML Tests .................... 3.88 |===============================
MBP M1 Max Machine Learning . 5.30 |===========================================
NCNN 20210720
Target: Vulkan GPU - Model: mobilenet
ms < Lower Is Better
ML Tests .................... 10.27 |=====================
MBP M1 Max Machine Learning . 20.30 |==========================================
Caffe 2020-02-13
Model: GoogleNet - Acceleration: CPU - Iterations: 100
Milli-Seconds < Lower Is Better
ML Tests . 86567 |=============================================================
OpenCV 4.5.4
Test: DNN - Deep Neural Network
ms < Lower Is Better
ML Tests . 13787 |=============================================================
Caffe 2020-02-13
Model: AlexNet - Acceleration: CPU - Iterations: 200
Milli-Seconds < Lower Is Better
ML Tests . 65986 |=============================================================
TensorFlow Lite 2020-08-23
Model: SqueezeNet
Microseconds < Lower Is Better
ML Tests . 189764 |============================================================
TensorFlow Lite 2020-08-23
Model: NASNet Mobile
Microseconds < Lower Is Better
ML Tests . 152186 |============================================================
TensorFlow Lite 2020-08-23
Model: Mobilenet Quant
Microseconds < Lower Is Better
ML Tests . 141174 |============================================================
TensorFlow Lite 2020-08-23
Model: Mobilenet Float
Microseconds < Lower Is Better
ML Tests . 127818 |============================================================
Mlpack Benchmark
Benchmark: scikit_ica
Seconds < Lower Is Better
ML Tests . 48.40 |=============================================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 3577.00 |========================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 3587.17 |========================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 3579.00 |========================================
Mlpack Benchmark
Benchmark: scikit_linearridgeregression
Seconds < Lower Is Better
ML Tests . 2.10 |==============================================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2228.17 |========================================
oneDNN 2.1.2
Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2219.13 |========================================
Caffe 2020-02-13
Model: AlexNet - Acceleration: CPU - Iterations: 100
Milli-Seconds < Lower Is Better
ML Tests . 33496 |=============================================================
DeepSpeech 0.6
Acceleration: CPU
Seconds < Lower Is Better
ML Tests .................... 74.44 |==========================================
Mlpack Benchmark
Benchmark: scikit_svm
Seconds < Lower Is Better
ML Tests . 17.60 |=============================================================
R Benchmark
Seconds < Lower Is Better
ML Tests . 0.1293 |============================================================
oneDNN 2.1.2
Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 4.25855 |========================================
TNN 0.3
Target: CPU - Model: MobileNet v2
ms < Lower Is Better
ML Tests . 249.48 |============================================================
RNNoise 2020-06-28
Seconds < Lower Is Better
ML Tests . 16.14 |=============================================================
TNN 0.3
Target: CPU - Model: SqueezeNet v1.1
ms < Lower Is Better
ML Tests . 222.33 |============================================================
ECP-CANDLE 0.4
Benchmark: P1B2
Seconds < Lower Is Better
ML Tests . 37.51 |=============================================================
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 8.34789 |========================================
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2.11771 |========================================
oneDNN 2.1.2
Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 1.62985 |========================================
oneDNN 2.1.2
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 4.59343 |========================================
oneDNN 2.1.2
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2.98659 |========================================
oneDNN 2.1.2
Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 12.09 |==========================================
oneDNN 2.1.2
Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 2.69210 |========================================
TNN 0.3
Target: CPU - Model: SqueezeNet v2
ms < Lower Is Better
ML Tests . 55.43 |=============================================================
oneDNN 2.1.2
Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 23.77 |==========================================
oneDNN 2.1.2
Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 22.79 |==========================================
Tensorflow
Build: Cifar10
Seconds < Lower Is Better
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 6.74559 |========================================
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU
ms < Lower Is Better
ML Tests .................... 3.24784 |========================================
Numenta Anomaly Benchmark 1.1
Detector: EXPoSE
Seconds < Lower Is Better
Numenta Anomaly Benchmark 1.1
Detector: Bayesian Changepoint
Seconds < Lower Is Better
Numenta Anomaly Benchmark 1.1
Detector: Relative Entropy
Seconds < Lower Is Better
Numenta Anomaly Benchmark 1.1
Detector: Earthgecko Skyline
Seconds < Lower Is Better
Numenta Anomaly Benchmark 1.1
Detector: Windowed Gaussian
Seconds < Lower Is Better
AI Benchmark Alpha 0.1.2
Score > Higher Is Better
ONNX Runtime 1.10
Model: yolov4 - Device: CPU
Inferences Per Minute > Higher Is Better
ONNX Runtime 1.10
Model: super-resolution-10 - Device: CPU
Inferences Per Minute > Higher Is Better
ONNX Runtime 1.10
Model: shufflenet-v2-10 - Device: CPU
Inferences Per Minute > Higher Is Better
ONNX Runtime 1.10
Model: fcn-resnet101-11 - Device: CPU
Inferences Per Minute > Higher Is Better
oneDNN 2.1.2
Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
oneDNN 2.1.2
Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
OpenVINO 2021.1
Model: Age Gender Recognition Retail 0013 FP32 - Device: Intel GPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Age Gender Recognition Retail 0013 FP16 - Device: Intel GPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Person Detection 0106 FP16 - Device: Intel GPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Face Detection 0106 FP32 - Device: CPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Face Detection 0106 FP16 - Device: CPU
FPS > Higher Is Better
oneDNN 2.1.2
Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
oneDNN 2.1.2
Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
oneDNN 2.1.2
Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU
ms < Lower Is Better
OpenVINO 2021.1
Model: Age Gender Recognition Retail 0013 FP32 - Device: CPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Person Detection 0106 FP32 - Device: Intel GPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Person Detection 0106 FP32 - Device: CPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Person Detection 0106 FP16 - Device: CPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Face Detection 0106 FP32 - Device: Intel GPU
FPS > Higher Is Better
OpenVINO 2021.1
Model: Face Detection 0106 FP16 - Device: Intel GPU
FPS > Higher Is Better