PlaidML OpenCL Linux GPU Benchmarks AMD Radeon + NVIDIA GeForce

PlaidML Linux benchmarks for a future article on Phoronix.com by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/1901122-PTS-PLAIDMLG16&sro.

PlaidML OpenCL Linux GPU Benchmarks AMD Radeon + NVIDIA GeForceProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionVulkanRX 580RX 590RX Vega 56RX Vega 64GTX 980GTX 980 TiGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTXAMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads)ASUS ROG ZENITH EXTREME (1601 BIOS)AMD Family 17h32768MBSamsung SSD 970 EVO 500GBMSI AMD Radeon RX 470/480/570/570X/580/580X 8GB (1366/2000MHz)Realtek ALC1220ASUS VP28UIntel I211 Gigabit + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11adUbuntu 18.104.19.0-041900-generic (x86_64)GNOME Shell 3.30.1X Server 1.20.1modesetting 1.20.14.5 Mesa 18.2.2 (LLVM 7.0.0)GCC 8.2.0ext43840x2160Sapphire AMD Radeon RX 470/480/570/570X/580/580X 8GB (1560/2100MHz)4.20.0-042000-generic (x86_64)AMD Radeon RX 64 8GB (1590/800MHz)4.19.0-041900-generic (x86_64)amdgpu 18.1.0AMD Radeon RX 64 8GB (1630/945MHz)NVIDIA GeForce GTX 980 4GB (1126/3505MHz)Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad4.20.0-042000-generic (x86_64)NVIDIA 415.254.6.01.1.84NVIDIA GeForce GTX 980 Ti 6GB (999/3505MHz)NVIDIA GeForce GTX 1060 6GB (1506/4006MHz)NVIDIA GeForce GTX 1070 8GB (1506/4006MHz)Zotac NVIDIA GeForce GTX 1070 Ti 8GB (1607/4006MHz)NVIDIA GeForce GTX 1080 8GB (1607/5005MHz)NVIDIA GeForce GTX 1080 Ti 11GB (1480/5508MHz)Device 6GB (1365/7000MHz)ASUS NVIDIA GeForce RTX 2070 8GB (1410/7000MHz)Zotac NVIDIA GeForce RTX 2080 8GB (1515/7000MHz)NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)Intel I211 Gigabit + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11adNVIDIA TITAN RTX 24GB (1350/7000MHz)Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11adOpenBenchmarking.orgProcessor Details- Scaling Governor: acpi-cpufreq ondemandGraphics Details- RX 580, RX 590, RX Vega 56, RX Vega 64: GLAMORPython Details- Python 2.7.15+ + Python 3.6.7Security Details- RX 580: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp- RX 590: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- RX Vega 56: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp- RX Vega 64: __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp- GTX 980: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 980 Ti: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 1060: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 1070: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 1070 Ti: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 1080: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- GTX 1080 Ti: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- RTX 2060: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- RTX 2070: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- RTX 2080: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- RTX 2080 Ti: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp- TITAN RTX: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccompOpenCL Details- GTX 980: GPU Compute Cores: 2048- GTX 980 Ti: GPU Compute Cores: 2816- GTX 1060: GPU Compute Cores: 1280- GTX 1070: GPU Compute Cores: 1920- GTX 1070 Ti: GPU Compute Cores: 2432- GTX 1080: GPU Compute Cores: 2560- GTX 1080 Ti: GPU Compute Cores: 3584- RTX 2060: GPU Compute Cores: 1920- RTX 2070: GPU Compute Cores: 2304- RTX 2080: GPU Compute Cores: 2944- RTX 2080 Ti: GPU Compute Cores: 4352- TITAN RTX: GPU Compute Cores: 4608

PlaidML OpenCL Linux GPU Benchmarks AMD Radeon + NVIDIA GeForceplaidml: No - Training - VGG16 - OpenCLplaidml: No - Training - VGG19 - OpenCLplaidml: No - Inference - VGG16 - OpenCLplaidml: No - Inference - VGG19 - OpenCLplaidml: Yes - Inference - VGG16 - OpenCLplaidml: Yes - Inference - VGG19 - OpenCLplaidml: No - Training - IMDB LSTM - OpenCLplaidml: No - Training - Mobilenet - OpenCLplaidml: No - Inference - IMDB LSTM - OpenCLplaidml: No - Inference - Mobilenet - OpenCLplaidml: No - Inference - ResNet 50 - OpenCLplaidml: Yes - Inference - Mobilenet - OpenCLplaidml: Yes - Inference - ResNet 50 - OpenCLplaidml: No - Training - Inception V3 - OpenCLplaidml: No - Inference - DenseNet 201 - OpenCLplaidml: No - Inference - Inception V3 - OpenCLplaidml: No - Inference - NASNer Large - OpenCLplaidml: Yes - Inference - DenseNet 201 - OpenCLplaidml: Yes - Inference - Inception V3 - OpenCLplaidml: Yes - Inference - NASNer Large - OpenCLRX 580RX 590RX Vega 56RX Vega 64GTX 980GTX 980 TiGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTX10.429.6764.5555.2658.9251.5911941.6418239013143512113.1562.3466.1918.089.528.9267.7158.3161.6553.8513345.2120542413945512914.5967.5071.4220.0128.7825.6610994.4110791.3216471.2924053121348019219.8695.8710431.0830.8528.15123107.32120103.9316975.2324054622648121220.98106.4411835.8994.8480.3796.8283.1910878.4715545817240717563.9681.0318.8961.3889.2623.2240.14119.20101.40121.68104.7412283.6617448519942119978.8896.6924.3276.15104.1026.9034.0793.0078.1392.1979.2911981.6517548518742617321.5862.2782.0318.2758.7686.7522.1745.9739.22125.37106.35122.54105.7713598.4521859324147622126.2183.89108.5025.3179.08112.3130.2046.7340.33128.85109.40125.60108.8513395.5221559424947322626.0083.55113.6126.5778.45114.3332.6550.0942.92138.78117.24134.68115.9014294.4724762127551024828.4490.83124.6530.9285.6013037.0266.5557.60181.25156.28180.71159.02165113.5529868731555531133.10119.80158.2042.75114.97160.1847.63147.16124.18143.80121.98157103.3032467129353825995.88124.8429.4289.0212637.6063.7154.76183.95155.08172.84148.93164119.8435676735153331131.44115.65147.3936.0110814244.3467.5958.55197.52168.53185.76161.16175120.2441879437361233032.80118.65161.9441.5711115351.5485.4574.75258.50221.88236.13207.30192138.1152596247869541737.67150.01197.5656.7713919269.7090.8479.58274.89237.71250.30220.89191142.3653598849969842837.41161.32211.6959.1314820372.61OpenBenchmarking.org

PlaidML

FP16: No - Mode: Training - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: VGG16 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980 TiRTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX20406080100SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.28, N = 3SE +/- 0.21, N = 3SE +/- 0.26, N = 3SE +/- 0.13, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.18, N = 334.0745.9746.7350.0966.5540.1463.7167.5985.4510.429.5228.7830.8590.84

PlaidML

FP16: No - Mode: Training - Network: VGG19 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: VGG19 - Device: OpenCLGTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX20406080100SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.19, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 339.2240.3342.9257.6054.7658.5574.759.678.9225.6628.1579.58

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX60120180240300SE +/- 0.06, N = 3SE +/- 0.40, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.45, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 3SE +/- 0.18, N = 3SE +/- 0.03, N = 3SE +/- 0.14, N = 3SE +/- 0.25, N = 3SE +/- 0.03, N = 3SE +/- 0.51, N = 3SE +/- 0.77, N = 3SE +/- 1.73, N = 3SE +/- 1.45, N = 393.00125.37128.85138.78181.2594.84119.20147.16183.95197.52258.5064.5567.71109.00123.00274.89

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX50100150200250SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.15, N = 3SE +/- 0.06, N = 3SE +/- 0.34, N = 3SE +/- 1.43, N = 4SE +/- 0.89, N = 3SE +/- 0.16, N = 378.13106.35109.40117.24156.2880.37101.40124.18155.08168.53221.8855.2658.3194.41107.32237.71

PlaidML

FP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG16 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX50100150200250SE +/- 0.82, N = 3SE +/- 0.04, N = 3SE +/- 0.36, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 1.48, N = 3SE +/- 0.06, N = 3SE +/- 0.27, N = 3SE +/- 0.06, N = 3SE +/- 0.39, N = 3SE +/- 0.14, N = 392.19122.54125.60134.68180.7196.82121.68143.80172.84185.76236.1358.9261.65107.00120.00250.30

PlaidML

FP16: Yes - Mode: Inference - Network: VGG19 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: VGG19 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX50100150200250SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 3SE +/- 1.40, N = 3SE +/- 0.15, N = 3SE +/- 0.17, N = 3SE +/- 0.21, N = 3SE +/- 0.06, N = 3SE +/- 0.38, N = 3SE +/- 0.06, N = 3SE +/- 0.71, N = 3SE +/- 0.04, N = 379.29105.77108.85115.90159.0283.19104.74121.98148.93161.16207.3051.5953.8591.32103.93220.89

PlaidML

FP16: No - Mode: Training - Network: IMDB LSTM - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: IMDB LSTM - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX4080120160200SE +/- 0.30, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.45, N = 3SE +/- 0.42, N = 3SE +/- 0.08, N = 3SE +/- 0.66, N = 3SE +/- 0.59, N = 3SE +/- 1.14, N = 3SE +/- 1.03, N = 3SE +/- 1.21, N = 3SE +/- 1.28, N = 3SE +/- 1.06, N = 3SE +/- 0.44, N = 3SE +/- 0.49, N = 3SE +/- 1.98, N = 3119135133142165108122157164175192119133164169191

PlaidML

FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: Mobilenet - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX306090120150SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.19, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.24, N = 3SE +/- 0.32, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 3SE +/- 0.40, N = 381.6598.4595.5294.47113.5578.4783.66103.30119.84120.24138.1141.6445.2171.2975.23142.36

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX120240360480600SE +/- 0.26, N = 3SE +/- 0.54, N = 3SE +/- 0.09, N = 3SE +/- 0.99, N = 3SE +/- 0.26, N = 3SE +/- 0.37, N = 3SE +/- 0.33, N = 3SE +/- 0.23, N = 3SE +/- 1.24, N = 3SE +/- 1.05, N = 3SE +/- 1.04, N = 3SE +/- 0.68, N = 3SE +/- 1.28, N = 3SE +/- 0.14, N = 3SE +/- 0.02, N = 3SE +/- 0.33, N = 3175218215247298155174324356418525182205240240535

PlaidML

FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX2004006008001000SE +/- 0.64, N = 3SE +/- 2.92, N = 3SE +/- 0.56, N = 3SE +/- 0.30, N = 3SE +/- 4.02, N = 3SE +/- 1.99, N = 3SE +/- 2.78, N = 3SE +/- 1.48, N = 3SE +/- 2.10, N = 3SE +/- 3.51, N = 3SE +/- 9.03, N = 3SE +/- 0.73, N = 3SE +/- 0.16, N = 3SE +/- 5.93, N = 3SE +/- 11.08, N = 12SE +/- 1.87, N = 3485593594621687458485671767794962390424531546988

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX110220330440550SE +/- 0.15, N = 3SE +/- 1.62, N = 3SE +/- 0.51, N = 3SE +/- 0.28, N = 3SE +/- 0.05, N = 3SE +/- 0.53, N = 3SE +/- 0.34, N = 3SE +/- 0.35, N = 3SE +/- 0.35, N = 3SE +/- 0.08, N = 3SE +/- 4.66, N = 3SE +/- 0.49, N = 3SE +/- 0.64, N = 3SE +/- 1.86, N = 3SE +/- 4.02, N = 3SE +/- 0.22, N = 3187241249275315172199293351373478131139213226499

PlaidML

FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX150300450600750SE +/- 5.36, N = 3SE +/- 1.57, N = 3SE +/- 8.00, N = 4SE +/- 4.10, N = 3SE +/- 5.47, N = 3SE +/- 4.97, N = 3SE +/- 2.96, N = 3SE +/- 7.50, N = 6SE +/- 6.98, N = 3SE +/- 6.97, N = 12SE +/- 1.64, N = 3SE +/- 1.39, N = 3SE +/- 0.50, N = 3SE +/- 0.30, N = 3SE +/- 0.64, N = 3SE +/- 5.84, N = 3426476473510555407421538533612695435455480481698

PlaidML

FP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: ResNet 50 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX90180270360450SE +/- 2.69, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 3SE +/- 0.31, N = 3SE +/- 0.73, N = 3SE +/- 0.19, N = 3SE +/- 0.23, N = 3SE +/- 6.28, N = 11SE +/- 0.17, N = 3SE +/- 0.43, N = 3SE +/- 0.30, N = 3SE +/- 0.32, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.88, N = 3SE +/- 0.23, N = 3173221226248311175199259311330417121129192212428

PlaidML

FP16: No - Mode: Training - Network: Inception V3 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Training - Network: Inception V3 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiRTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX918273645SE +/- 0.10, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.29, N = 321.5826.2126.0028.4433.1031.4432.8037.6713.1514.5919.8620.9837.41

PlaidML

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX4080120160200SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.13, N = 3SE +/- 2.09, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.42, N = 362.2783.8983.5590.83119.8063.9678.8895.88115.65118.65150.0162.3467.5095.87106.44161.32

PlaidML

FP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Inception V3 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX50100150200250SE +/- 1.04, N = 7SE +/- 1.68, N = 5SE +/- 0.05, N = 3SE +/- 1.75, N = 12SE +/- 0.42, N = 3SE +/- 1.08, N = 3SE +/- 1.44, N = 5SE +/- 1.89, N = 3SE +/- 0.61, N = 3SE +/- 0.12, N = 3SE +/- 3.98, N = 12SE +/- 0.95, N = 3SE +/- 1.29, N = 3SE +/- 1.56, N = 3SE +/- 0.77, N = 3SE +/- 0.31, N = 382.03108.50113.61124.65158.2081.0396.69124.84147.39161.94197.5666.1971.42104.00118.00211.69

PlaidML

FP16: No - Mode: Inference - Network: NASNer Large - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: NASNer Large - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiRX 580RX 590RX Vega 56RX Vega 64TITAN RTX1326395265SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 318.2725.3126.5730.9242.7518.8924.3229.4236.0141.5756.7718.0820.0131.0835.8959.13

PlaidML

FP16: Yes - Mode: Inference - Network: DenseNet 201 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: DenseNet 201 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTX306090120150SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.18, N = 3SE +/- 0.17, N = 3SE +/- 0.10, N = 3SE +/- 1.79, N = 7SE +/- 2.16, N = 358.7679.0878.4585.60114.9761.3876.1589.02108.00111.00139.00148.00

PlaidML

FP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: Inception V3 - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTX4080120160200SE +/- 0.27, N = 3SE +/- 0.16, N = 3SE +/- 1.45, N = 7SE +/- 0.17, N = 3SE +/- 0.55, N = 3SE +/- 0.15, N = 3SE +/- 1.04, N = 12SE +/- 2.35, N = 3SE +/- 2.27, N = 12SE +/- 2.75, N = 12SE +/- 3.75, N = 12SE +/- 0.41, N = 386.75112.31114.33130.00160.1889.26104.10126.00142.00153.00192.00203.00

PlaidML

FP16: Yes - Mode: Inference - Network: NASNer Large - Device: OpenCL

OpenBenchmarking.orgExamples Per Second, More Is BetterPlaidMLFP16: Yes - Mode: Inference - Network: NASNer Large - Device: OpenCLGTX 1060GTX 1070GTX 1070 TiGTX 1080GTX 1080 TiGTX 980GTX 980 TiRTX 2060RTX 2070RTX 2080RTX 2080 TiTITAN RTX1632486480SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.89, N = 3SE +/- 0.16, N = 322.1730.2032.6537.0247.6323.2226.9037.6044.3451.5469.7072.61

Phoronix Test Suite v10.8.4