Jetson Nano Developer Kit Benchmarks for a future article on Phoronix.com. Jetson TX1 Max-P: Processor: ARMv8 rev 1 @ 1.73GHz (4 Cores), Motherboard: jetson_tx1, Memory: 4096MB, Disk: 16GB 016G32, Graphics: NVIDIA Tegra X1, Monitor: VE228 OS: Ubuntu 16.04, Kernel: 4.4.38-tegra (aarch64), Desktop: Unity 7.4.5, Display Server: X Server 1.18.4, Display Driver: NVIDIA 28.1.0, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.4.0 20160609, File-System: ext4, Screen Resolution: 1920x1080 Jetson TX2 Max-Q: Processor: ARMv8 rev 3 @ 1.27GHz (4 Cores / 6 Threads), Motherboard: quill, Memory: 8192MB, Disk: 31GB 032G34, Graphics: NVIDIA TEGRA, Monitor: VE228 OS: Ubuntu 16.04, Kernel: 4.4.38-tegra (aarch64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 28.2.1, OpenGL: 4.5.0, Compiler: GCC 5.4.0 20160609 + CUDA 9.0, File-System: ext4, Screen Resolution: 1920x1080 Jetson TX2 Max-P: Processor: ARMv8 rev 3 @ 2.04GHz (4 Cores / 6 Threads), Motherboard: quill, Memory: 8192MB, Disk: 31GB 032G34, Graphics: NVIDIA TEGRA, Monitor: VE228 OS: Ubuntu 16.04, Kernel: 4.4.38-tegra (aarch64), Desktop: Unity 7.4.0, Display Server: X Server 1.18.4, Display Driver: NVIDIA 28.2.1, OpenGL: 4.5.0, Compiler: GCC 5.4.0 20160609 + CUDA 9.0, File-System: ext4, Screen Resolution: 1920x1080 Jetson AGX Xavier: Processor: ARMv8 rev 0 @ 2.27GHz (8 Cores), Motherboard: jetson-xavier, Memory: 16384MB, Disk: 31GB HBG4a2, Graphics: NVIDIA Tegra Xavier, Monitor: VE228 OS: Ubuntu 18.04, Kernel: 4.9.108-tegra (aarch64), Desktop: Unity 7.5.0, Display Server: X Server 1.19.6, Display Driver: NVIDIA 31.0.2, OpenGL: 4.6.0, Vulkan: 1.1.76, Compiler: GCC 7.3.0 + CUDA 10.0, File-System: ext4, Screen Resolution: 1920x1080 Jetson Nano: Processor: ARMv8 rev 1 @ 1.43GHz (4 Cores), Motherboard: jetson-nano, Memory: 4096MB, Disk: 32GB GB1QT, Graphics: NVIDIA TEGRA, Monitor: VE228, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 18.04, Kernel: 4.9.140-tegra (aarch64), Desktop: Unity 7.5.0, Display Server: X Server 1.19.6, Display Driver: NVIDIA 1.0.0, Vulkan: 1.1.85, Compiler: GCC 7.3.0 + CUDA 10.0, File-System: ext4, Screen Resolution: 1920x1080 Raspberry Pi 3 Model B+: Processor: ARMv7 rev 4 @ 1.40GHz (4 Cores), Motherboard: BCM2835 Raspberry Pi 3 Model B Plus Rev 1.3, Memory: 926MB, Disk: 32GB GB2MW, Graphics: BCM2708 OS: Raspbian 9.6, Kernel: 4.19.23-v7+ (armv7l), Desktop: LXDE, Display Server: X Server 1.19.2, Compiler: GCC 6.3.0 20170516, File-System: ext4, Screen Resolution: 656x416 ASUS TinkerBoard: Processor: ARMv7 rev 1 @ 1.80GHz (4 Cores), Motherboard: Rockchip (Device Tree), Memory: 2048MB, Disk: 32GB GB1QT OS: Debian 9.0, Kernel: 4.4.16-00006-g4431f98-dirty (armv7l), Desktop: LXDE, Display Server: X Server 1.18.4, Compiler: GCC 6.3.0 20170516, File-System: ext4, Screen Resolution: 1024x768 ODROID-XU4: Processor: ARMv7 rev 3 @ 1.50GHz (8 Cores), Motherboard: ODROID-XU4 Hardkernel Odroid XU4, Memory: 2048MB, Disk: 16GB AJTD4R, Graphics: llvmpipe 2GB, Monitor: VE228 OS: Ubuntu 18.04, Kernel: 4.14.37-135 (armv7l), Display Server: X Server 1.19.6, OpenGL: 3.3 Mesa 18.0.0-rc5 (LLVM 6.0 128 bits), Compiler: GCC 7.3.0, File-System: ext4, Screen Resolution: 1920x1080 CUDA Mini-Nbody 2015-11-10 Test: Original (NBody^2)/s > Higher Is Better Jetson TX2 Max-Q .. 6.77 |======= Jetson TX2 Max-P .. 8.24 |========= Jetson AGX Xavier . 47.13 |==================================================== Jetson Nano ....... 4.07 |==== GLmark2 Resolution: 1920 x 1080 Score > Higher Is Better Jetson AGX Xavier . 2876 |===================================================== Jetson Nano ....... 646 |============ NVIDIA TensorRT Inference Neural Network: VGG16 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 25.99 |====== Jetson TX2 Max-P .. 32.64 |======== Jetson AGX Xavier . 208.76 |=================================================== Jetson Nano ....... 14.35 |==== NVIDIA TensorRT Inference Neural Network: VGG16 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 14.24 |== Jetson TX2 Max-P .. 17.56 |=== Jetson AGX Xavier . 303.78 |=================================================== NVIDIA TensorRT Inference Neural Network: VGG19 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 21.04 |====== Jetson TX2 Max-P .. 26.56 |======== Jetson AGX Xavier . 172.50 |=================================================== Jetson Nano ....... 11.59 |=== NVIDIA TensorRT Inference Neural Network: VGG19 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 11.45 |== Jetson TX2 Max-P .. 14.32 |=== Jetson AGX Xavier . 265.81 |=================================================== NVIDIA TensorRT Inference Neural Network: VGG16 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 29.83 |====== Jetson TX2 Max-P .. 36.87 |======== Jetson AGX Xavier . 247.95 |=================================================== NVIDIA TensorRT Inference Neural Network: VGG16 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 15.79 |== Jetson TX2 Max-P .. 19.91 |== Jetson AGX Xavier . 475.08 |=================================================== NVIDIA TensorRT Inference Neural Network: VGG19 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 23.94 |====== Jetson TX2 Max-P .. 29.83 |======= Jetson AGX Xavier . 203.96 |=================================================== NVIDIA TensorRT Inference Neural Network: VGG19 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 12.59 |== Jetson TX2 Max-P .. 15.92 |== Jetson AGX Xavier . 394.66 |=================================================== NVIDIA TensorRT Inference Neural Network: AlexNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 216 |========== Jetson TX2 Max-P .. 264 |============ Jetson AGX Xavier . 1200 |===================================================== Jetson Nano ....... 118 |===== NVIDIA TensorRT Inference Neural Network: AlexNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 148.00 |====== Jetson TX2 Max-P .. 184.00 |======== Jetson AGX Xavier . 1143.00 |================================================== Jetson Nano ....... 84.10 |==== NVIDIA TensorRT Inference Neural Network: AlexNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 374 |========== Jetson TX2 Max-P .. 462 |============ Jetson AGX Xavier . 2038 |===================================================== Jetson Nano ....... 201 |===== NVIDIA TensorRT Inference Neural Network: AlexNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 237 |==== Jetson TX2 Max-P .. 301 |===== Jetson AGX Xavier . 3143 |===================================================== Jetson Nano ....... 128 |== NVIDIA TensorRT Inference Neural Network: ResNet50 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 72.01 |======= Jetson TX2 Max-P .. 92.28 |========= Jetson AGX Xavier . 547.50 |=================================================== Jetson Nano ....... 41.04 |==== NVIDIA TensorRT Inference Neural Network: ResNet50 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 39.15 |== Jetson TX2 Max-P .. 49.97 |=== Jetson AGX Xavier . 902.78 |=================================================== Jetson Nano ....... 20.96 |= NVIDIA TensorRT Inference Neural Network: GoogleNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 156.00 |========== Jetson TX2 Max-P .. 197.00 |============= Jetson AGX Xavier . 796.00 |=================================================== Jetson Nano ....... 83.37 |===== NVIDIA TensorRT Inference Neural Network: GoogleNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 88.88 |==== Jetson TX2 Max-P .. 113.00 |===== Jetson AGX Xavier . 1146.00 |================================================== Jetson Nano ....... 47.82 |== NVIDIA TensorRT Inference Neural Network: ResNet152 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 27.34 |====== Jetson TX2 Max-P .. 35.11 |======== Jetson AGX Xavier . 224.19 |=================================================== Jetson Nano ....... 15.76 |==== NVIDIA TensorRT Inference Neural Network: ResNet152 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 14.50 |== Jetson TX2 Max-P .. 18.29 |=== Jetson AGX Xavier . 372.73 |=================================================== Jetson Nano ....... 7.76 |= NVIDIA TensorRT Inference Neural Network: ResNet50 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 86.08 |======= Jetson TX2 Max-P .. 111.00 |========= Jetson AGX Xavier . 636.00 |=================================================== Jetson Nano ....... 46.51 |==== NVIDIA TensorRT Inference Neural Network: ResNet50 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 47.15 |== Jetson TX2 Max-P .. 59.69 |== Jetson AGX Xavier . 1215.08 |================================================== Jetson Nano ....... 25.08 |= NVIDIA TensorRT Inference Neural Network: GoogleNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 179.00 |========= Jetson TX2 Max-P .. 233.00 |============ Jetson AGX Xavier . 1006.00 |================================================== Jetson Nano ....... 98.93 |===== NVIDIA TensorRT Inference Neural Network: GoogleNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 104.00 |=== Jetson TX2 Max-P .. 130.00 |==== Jetson AGX Xavier . 1693.00 |================================================== Jetson Nano ....... 55.66 |== NVIDIA TensorRT Inference Neural Network: ResNet152 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 32.67 |====== Jetson TX2 Max-P .. 41.91 |======== Jetson AGX Xavier . 259.82 |=================================================== Jetson Nano ....... 17.38 |=== NVIDIA TensorRT Inference Neural Network: ResNet152 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second > Higher Is Better Jetson TX2 Max-Q .. 17.36 |== Jetson TX2 Max-P .. 22.07 |== Jetson AGX Xavier . 493.22 |=================================================== LeelaChessZero 0.20.1 Backend: BLAS Nodes Per Second > Higher Is Better Jetson AGX Xavier . 47.62 |==================================================== Jetson Nano ....... 15.37 |================= LeelaChessZero 0.20.1 Backend: CUDA + cuDNN Nodes Per Second > Higher Is Better Jetson AGX Xavier . 953 |====================================================== Jetson Nano ....... 140 |======== LeelaChessZero 0.20.1 Backend: CUDA + cuDNN FP16 Nodes Per Second > Higher Is Better Jetson AGX Xavier . 2515.01 |================================================== TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping FPS > Higher Is Better Jetson TX1 Max-P ........ 45.09 |=============== Jetson TX2 Max-Q ........ 28.85 |========== Jetson TX2 Max-P ........ 49.26 |================= Jetson AGX Xavier ....... 133.00 |============================================= Jetson Nano ............. 40.94 |============== Raspberry Pi 3 Model B+ . 17.66 |====== ASUS TinkerBoard ........ 21.22 |======= ODROID-XU4 .............. 41.96 |============== 7-Zip Compression 16.02 Compress Speed Test MIPS > Higher Is Better Jetson TX1 Max-P ........ 4508 |=========== Jetson TX2 Max-Q ........ 3294 |======== Jetson TX2 Max-P ........ 5593 |============= Jetson AGX Xavier ....... 19212 |============================================== Jetson Nano ............. 4049 |========== Raspberry Pi 3 Model B+ . 2013 |===== ASUS TinkerBoard ........ 2836 |======= ODROID-XU4 .............. 4120 |========== C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better Jetson TX1 Max-P ........ 753 |================= Jetson TX2 Max-Q ........ 869 |==================== Jetson TX2 Max-P ........ 585 |============== Jetson AGX Xavier ....... 355 |======== Jetson Nano ............. 921 |===================== Raspberry Pi 3 Model B+ . 2030 |=============================================== ASUS TinkerBoard ........ 1718 |======================================== ODROID-XU4 .............. 827 |=================== Rust Prime Benchmark Prime Number Test To 200,000,000 Seconds < Lower Is Better Jetson TX1 Max-P ........ 128.45 |=== Jetson TX2 Max-Q ........ 170.25 |==== Jetson TX2 Max-P ........ 104.96 |=== Jetson AGX Xavier ....... 32.37 |= Jetson Nano ............. 150.19 |==== Raspberry Pi 3 Model B+ . 1097.69 |=========================== ASUS TinkerBoard ........ 1821.05 |============================================ ODROID-XU4 .............. 574.11 |============== Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds < Lower Is Better Jetson TX1 Max-P ........ 145.80 |============= Jetson TX2 Max-Q ........ 253.80 |======================= Jetson TX2 Max-P ........ 144.97 |============= Jetson AGX Xavier ....... 80.06 |======= Jetson Nano ............. 129.87 |============ Raspberry Pi 3 Model B+ . 342.23 |=============================== ASUS TinkerBoard ........ 496.62 |============================================= FLAC Audio Encoding 1.3.2 WAV To FLAC Seconds < Lower Is Better Jetson TX1 Max-P ........ 79.20 |========== Jetson TX2 Max-Q ........ 104.28 |============== Jetson TX2 Max-P ........ 65.07 |========= Jetson AGX Xavier ....... 54.47 |======= Jetson Nano ............. 104.77 |============== Raspberry Pi 3 Model B+ . 339.53 |============================================= ASUS TinkerBoard ........ 279.05 |===================================== ODROID-XU4 .............. 97.03 |============= OpenCV Benchmark 3.3.0 Seconds < Lower Is Better Jetson TX2 Max-Q ........ 493.00 |=========================================== Jetson TX2 Max-P ........ 296.00 |========================== Jetson AGX Xavier ....... 128.00 |=========== Jetson Nano ............. 271.04 |======================= Raspberry Pi 3 Model B+ . 2.74 | ODROID-XU4 .............. 520.70 |============================================= PyBench 2018-02-16 Total For Average Test Times Milliseconds < Lower Is Better Jetson TX1 Max-P ........ 6339 |============== Jetson TX2 Max-Q ........ 8735 |=================== Jetson TX2 Max-P ........ 5408 |============ Jetson AGX Xavier ....... 3007 |======= Jetson Nano ............. 7084 |================ Raspberry Pi 3 Model B+ . 20913 |============================================== ASUS TinkerBoard ........ 11502 |========================= ODROID-XU4 .............. 5009 |=========== Tesseract OCR 4.0.0-beta.1 Time To OCR 7 Images Seconds < Lower Is Better Jetson AGX Xavier . 71.94 |==================== Jetson Nano ....... 132.67 |===================================== ODROID-XU4 ........ 180.66 |=================================================== TTSIOD 3D Renderer 2.3b Performance / Cost - Phong Rendering With Soft-Shadow Mapping FPS Per Dollar > Higher Is Better Jetson TX1 Max-P ........ 0.09 |====== Jetson TX2 Max-Q ........ 0.05 |=== Jetson TX2 Max-P ........ 0.08 |====== Jetson AGX Xavier ....... 0.10 |======= Jetson Nano ............. 0.41 |============================ Raspberry Pi 3 Model B+ . 0.50 |=================================== ASUS TinkerBoard ........ 0.32 |====================== ODROID-XU4 .............. 0.68 |=============================================== 7-Zip Compression 16.02 Performance / Cost - Compress Speed Test MIPS Per Dollar > Higher Is Better Jetson TX1 Max-P ........ 9.03 |====== Jetson TX2 Max-Q ........ 5.50 |==== Jetson TX2 Max-P ........ 9.34 |====== Jetson AGX Xavier ....... 14.79 |========== Jetson Nano ............. 40.90 |============================ Raspberry Pi 3 Model B+ . 57.51 |======================================== ASUS TinkerBoard ........ 42.97 |============================== ODROID-XU4 .............. 66.45 |============================================== C-Ray 1.1 Performance / Cost - Total Time - 4K, 16 Rays Per Pixel Seconds x Dollar < Lower Is Better Jetson TX1 Max-P ........ 375747.00 |============================== Jetson TX2 Max-Q ........ 520531.00 |========================================== Jetson TX2 Max-P ........ 350415.00 |============================ Jetson AGX Xavier ....... 461145.00 |===================================== Jetson Nano ............. 91179.00 |======= Raspberry Pi 3 Model B+ . 71050.00 |====== ASUS TinkerBoard ........ 113388.00 |========= ODROID-XU4 .............. 51274.00 |==== Rust Prime Benchmark Performance / Cost - Prime Number Test To 200,000,000 Seconds x Dollar < Lower Is Better Jetson TX1 Max-P ........ 64096.55 |====================== Jetson TX2 Max-Q ........ 101979.75 |==================================== Jetson TX2 Max-P ........ 62871.04 |====================== Jetson AGX Xavier ....... 42048.63 |=============== Jetson Nano ............. 14868.81 |===== Raspberry Pi 3 Model B+ . 38419.15 |============= ASUS TinkerBoard ........ 120189.30 |========================================== ODROID-XU4 .............. 35594.82 |============ Zstd Compression 1.3.4 Performance / Cost - Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Seconds x Dollar < Lower Is Better Jetson TX1 Max-P ........ 72754.20 |==================== Jetson TX2 Max-Q ........ 152026.20 |========================================== Jetson TX2 Max-P ........ 86837.03 |======================== Jetson AGX Xavier ....... 103997.94 |============================= Jetson Nano ............. 12857.13 |==== Raspberry Pi 3 Model B+ . 11978.05 |=== ASUS TinkerBoard ........ 32776.92 |========= FLAC Audio Encoding 1.3.2 Performance / Cost - WAV To FLAC Seconds x Dollar < Lower Is Better Jetson TX1 Max-P ........ 39520.80 |======================== Jetson TX2 Max-Q ........ 62463.72 |====================================== Jetson TX2 Max-P ........ 38976.93 |======================== Jetson AGX Xavier ....... 70756.53 |=========================================== Jetson Nano ............. 10372.23 |====== Raspberry Pi 3 Model B+ . 11883.55 |======= ASUS TinkerBoard ........ 18417.30 |=========== ODROID-XU4 .............. 6015.86 |==== PyBench 2018-02-16 Performance / Cost - Total For Average Test Times Milliseconds x Dollar < Lower Is Better Jetson TX1 Max-P ........ 3163161.00 |========================= Jetson TX2 Max-Q ........ 5232265.00 |========================================= Jetson TX2 Max-P ........ 3239392.00 |========================= Jetson AGX Xavier ....... 3906093.00 |=============================== Jetson Nano ............. 701316.00 |===== Raspberry Pi 3 Model B+ . 731955.00 |====== ASUS TinkerBoard ........ 759132.00 |====== ODROID-XU4 .............. 310558.00 |== CUDA Mini-Nbody 2015-11-10 Performance / Cost - Test: Original (NBody^2)/s Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.01 |============= Jetson TX2 Max-P .. 0.01 |============= Jetson AGX Xavier . 0.04 |===================================================== Jetson Nano ....... 0.04 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG16 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.04 |============= Jetson TX2 Max-P .. 0.05 |================= Jetson AGX Xavier . 0.16 |===================================================== Jetson Nano ....... 0.14 |============================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG16 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.02 |===== Jetson TX2 Max-P .. 0.03 |======= Jetson AGX Xavier . 0.23 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG19 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.04 |================ Jetson TX2 Max-P .. 0.04 |================ Jetson AGX Xavier . 0.13 |===================================================== Jetson Nano ....... 0.12 |================================================= NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG19 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.02 |===== Jetson TX2 Max-P .. 0.02 |===== Jetson AGX Xavier . 0.20 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG16 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.05 |============== Jetson TX2 Max-P .. 0.06 |================= Jetson AGX Xavier . 0.19 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG16 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.03 |==== Jetson TX2 Max-P .. 0.03 |==== Jetson AGX Xavier . 0.37 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG19 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.04 |============= Jetson TX2 Max-P .. 0.05 |================= Jetson AGX Xavier . 0.16 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: VGG19 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.02 |==== Jetson TX2 Max-P .. 0.03 |===== Jetson AGX Xavier . 0.30 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: AlexNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.36 |================ Jetson TX2 Max-P .. 0.44 |==================== Jetson AGX Xavier . 0.92 |========================================= Jetson Nano ....... 1.19 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: AlexNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.25 |=============== Jetson TX2 Max-P .. 0.31 |=================== Jetson AGX Xavier . 0.88 |===================================================== Jetson Nano ....... 0.85 |=================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: AlexNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.62 |================ Jetson TX2 Max-P .. 0.77 |==================== Jetson AGX Xavier . 1.57 |========================================= Jetson Nano ....... 2.03 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: AlexNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.40 |========= Jetson TX2 Max-P .. 0.50 |=========== Jetson AGX Xavier . 2.42 |===================================================== Jetson Nano ....... 1.29 |============================ NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.12 |=============== Jetson TX2 Max-P .. 0.15 |=================== Jetson AGX Xavier . 0.42 |===================================================== Jetson Nano ....... 0.41 |==================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.07 |===== Jetson TX2 Max-P .. 0.08 |====== Jetson AGX Xavier . 0.69 |===================================================== Jetson Nano ....... 0.21 |================ NVIDIA TensorRT Inference Performance / Cost - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.26 |================ Jetson TX2 Max-P .. 0.33 |===================== Jetson AGX Xavier . 0.61 |====================================== Jetson Nano ....... 0.84 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.15 |========= Jetson TX2 Max-P .. 0.19 |=========== Jetson AGX Xavier . 0.88 |===================================================== Jetson Nano ....... 0.48 |============================= NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.05 |================ Jetson TX2 Max-P .. 0.06 |=================== Jetson AGX Xavier . 0.17 |===================================================== Jetson Nano ....... 0.16 |================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.02 |==== Jetson TX2 Max-P .. 0.03 |===== Jetson AGX Xavier . 0.29 |===================================================== Jetson Nano ....... 0.08 |=============== NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.14 |=============== Jetson TX2 Max-P .. 0.19 |===================== Jetson AGX Xavier . 0.49 |===================================================== Jetson Nano ....... 0.47 |=================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.08 |===== Jetson TX2 Max-P .. 0.10 |====== Jetson AGX Xavier . 0.94 |===================================================== Jetson Nano ....... 0.25 |============== NVIDIA TensorRT Inference Performance / Cost - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.30 |================ Jetson TX2 Max-P .. 0.39 |===================== Jetson AGX Xavier . 0.77 |========================================= Jetson Nano ....... 1.00 |===================================================== NVIDIA TensorRT Inference Performance / Cost - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.17 |======= Jetson TX2 Max-P .. 0.22 |========= Jetson AGX Xavier . 1.30 |===================================================== Jetson Nano ....... 0.56 |======================= NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.05 |============= Jetson TX2 Max-P .. 0.07 |=================== Jetson AGX Xavier . 0.20 |===================================================== Jetson Nano ....... 0.18 |================================================ NVIDIA TensorRT Inference Performance / Cost - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled Images Per Second Per Dollar > Higher Is Better Jetson TX2 Max-Q .. 0.03 |==== Jetson TX2 Max-P .. 0.04 |====== Jetson AGX Xavier . 0.38 |===================================================== OpenCV Benchmark 3.3.0 Performance / Cost - Seconds x Dollar < Lower Is Better Jetson TX2 Max-Q ........ 295307.00 |========================================== Jetson TX2 Max-P ........ 177304.00 |========================= Jetson AGX Xavier ....... 166272.00 |======================== Jetson Nano ............. 26832.96 |==== Raspberry Pi 3 Model B+ . 95.90 | ODROID-XU4 .............. 32283.40 |===== GLmark2 Performance / Cost - Resolution: 1920 x 1080 Score Per Dollar > Higher Is Better Jetson AGX Xavier . 2.21 |================== Jetson Nano ....... 6.53 |===================================================== LeelaChessZero 0.20.1 Performance / Cost - Backend: BLAS Nodes Per Second Per Dollar > Higher Is Better Jetson AGX Xavier . 0.04 |============= Jetson Nano ....... 0.16 |===================================================== LeelaChessZero 0.20.1 Performance / Cost - Backend: CUDA + cuDNN Nodes Per Second Per Dollar > Higher Is Better Jetson AGX Xavier . 0.73 |=========================== Jetson Nano ....... 1.41 |===================================================== LeelaChessZero 0.20.1 Performance / Cost - Backend: CUDA + cuDNN FP16 Nodes Per Second Per Dollar > Higher Is Better Jetson AGX Xavier . 1.94 |===================================================== Tesseract OCR 4.0.0-beta.1 Performance / Cost - Time To OCR 7 Images Seconds x Dollar < Lower Is Better Jetson AGX Xavier . 93450.06 |================================================= Jetson Nano ....... 13134.33 |======= ODROID-XU4 ........ 11200.92 |======