Benchmarks for a future article on Phoronix.com.
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 1908193-HV-1903186HV68
Jetson Nano Developer Kit,
"FLAC Audio Encoding 1.3.2 - WAV To FLAC",
Lower Results Are Better
"Jetson TX1 Max-P",80.643102169037,80.375013828278,80.03696680069,76.877398014069,78.043719053268
"Jetson TX2 Max-Q",103.99030089378,104.48815512657,104.76083803177,104.37327098846,103.7655339241
"Jetson TX2 Max-P",65.099822998047,65.641438961029,64.799404144287,64.830460071564,64.956172943115
"Jetson AGX Xavier",56.730622053146,54.393737077713,53.064939975739,53.975198030472,54.179378032684
"Jetson Nano",103.41150212288,105.25729489326,103.14293313026,107.75390219688,104.28678107262
"Raspberry Pi 3 Model B+",339.31037282944,337.10595822334,342.56921696663,340.70349907875,337.95009493828
"ASUS TinkerBoard",286.25495100021,281.78744792938,279.70542287827,271.50059986115,275.99261808395
"ODROID-XU4",97.058123111725,96.899234056473,96.224318027496,96.844074964523,98.121542930603
"AGX Xavier 32.2",51.546104192734,49.241894006729,49.701735019684,50.274689912796,50.474295854568
"Tesseract OCR 4.0.0-beta.1 - Time To OCR 7 Images",
Lower Results Are Better
"Jetson AGX Xavier",73.572254896164,71.729460000992,70.515511989594
"Jetson Nano",135.4518301487,132.27096486092,130.28921198845
"ODROID-XU4",180.56019186974,183.09073805809,178.31624317169
"AGX Xavier 32.2",90.315413951874,90.007643938065,90.092839002609
"NVIDIA TensorRT Inference - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",90.6201,86.2883,89.7201
"Jetson TX2 Max-P",110.896,112.155,116.361
"Jetson AGX Xavier",1137.32,1151.7,1148.02
"Jetson Nano",46.8307,48.891,47.7487
"AGX Xavier 32.2",1198.34,1191.02,1202.68
"LeelaChessZero 0.20.1 - Backend: BLAS",
Higher Results Are Better
"Jetson AGX Xavier",51.2966,47.2233,46.4651,47.1557,46.7206,47.1349,47.3134
"Jetson Nano",15.3987,15.3082,15.3961
"AGX Xavier 32.2",53.8943,53.7848,53.3368
"NVIDIA TensorRT Inference - Neural Network: AlexNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",377.966,368.297,374.434
"Jetson TX2 Max-P",418.289,485.379,493.491,432.306,488.111,441.301,465.489,479.41,435.698,473.94,440.839,489.368
"Jetson AGX Xavier",2037.23,2042.31,2035.4
"Jetson Nano",202.539,202.319,197.659
"AGX Xavier 32.2",2029.45,2025.37,2032.79
"LeelaChessZero 0.20.1 - Backend: CUDA + cuDNN",
Higher Results Are Better
"Jetson AGX Xavier",960.648,940.764,957.256
"Jetson Nano",140.308,139.466,139.604
"AGX Xavier 32.2",972.911,966.387,977.318
"NVIDIA TensorRT Inference - Neural Network: VGG19 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",20.8847,21.6977,20.5357
"Jetson TX2 Max-P",26.5966,27.2027,25.8709
"Jetson AGX Xavier",173.429,171.708,172.356
"Jetson Nano",11.6433,11.5434
"AGX Xavier 32.2",170.26,169.39,169.61
"NVIDIA TensorRT Inference - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",27.984,26.8478,27.1896
"Jetson TX2 Max-P",35.6056,34.4145,35.3045
"Jetson AGX Xavier",224.335,223.747,224.477
"Jetson Nano",15.8286,15.7086,15.7311
"AGX Xavier 32.2",231.483,232.634,231.511
"NVIDIA TensorRT Inference - Neural Network: VGG16 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",26.1732,26.0698,25.7346
"Jetson TX2 Max-P",32.143,31.6987,33.981,32.7477
"Jetson AGX Xavier",208.703,208.962,208.628
"Jetson Nano",14.3692,14.3311
"AGX Xavier 32.2",203.196,204.301,203.945
"NVIDIA TensorRT Inference - Neural Network: VGG19 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",24.037,23.7941,23.9907
"Jetson TX2 Max-P",29.7627,29.9376,29.7943
"Jetson AGX Xavier",203.915,203.924,204.032
"AGX Xavier 32.2",127.861,112.229,107.792,120.218,110.259,110.621,105.232,115.955,107.777,112.548,116.737,110.803,116.902,107.308,121.613
"NVIDIA TensorRT Inference - Neural Network: VGG16 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",29.4819,29.8775,30.1157
"Jetson TX2 Max-P",37.1824,36.2465,37.1919
"Jetson AGX Xavier",248.139,247.726,247.981
"AGX Xavier 32.2",192.353,173.252,170.71,169.728,169.646,155.948,157.528,156.833,169.224,157.416,155.715,155.619,162.546,159.646,165.923
"NVIDIA TensorRT Inference - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",153.162,160.497,165.878,151.554,163.542,148.371,155.992,155.301,148.849,161.288,163.564,146.96
"Jetson TX2 Max-P",193.491,201.347,197.376
"Jetson AGX Xavier",797.316,790.769,798.875
"Jetson Nano",82.7407,84.7738,82.6049
"AGX Xavier 32.2",861.881,862.084,870.282
"NVIDIA TensorRT Inference - Neural Network: AlexNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",202.585,224.229,215.21,217.911,218.577,220.182
"Jetson TX2 Max-P",278.36,222.341,249.778,255.787,277.473,304.014,257.594,228.296,294.245,245.232,299.846,253.799
"Jetson AGX Xavier",1201.46,1201.91,1196.24
"Jetson Nano",120.801,109.249,113.019,127.583,125.389,114.899,113.584,120.422,118.392,125.583,126.446,104.749
"AGX Xavier 32.2",1196.9,1193.62,1195.1
"NVIDIA TensorRT Inference - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",77.5964,74.5251,72.0198,66.272,69.865,75.839,74.3196,70.7943,69.4287,73.1548,65.2801,74.9964
"Jetson TX2 Max-P",94.1442,86.2995,95.9709,91.9219,95.8238,89.0835,89.939,89.334,85.3654,91.5064,100.018,97.9283
"Jetson AGX Xavier",547.485,547.458,547.56
"Jetson Nano",40.8261,41.5429,40.749
"AGX Xavier 32.2",589.042,590.61,590.287
"LeelaChessZero 0.20.1 - Backend: CUDA + cuDNN FP16",
Higher Results Are Better
"Jetson AGX Xavier",2523.19,2499.83,2522.01
"AGX Xavier 32.2",2514.1,2464.37,2579.75
"Rust Prime Benchmark - Prime Number Test To 200,000,000",
Lower Results Are Better
"Jetson TX1 Max-P",128.75703287125,129.60009694099,127.0019800663
"Jetson TX2 Max-Q",170.26361608505,170.08605980873,170.4050219059
"Jetson TX2 Max-P",104.92546510696,105.04112505913,104.90127086639
"Jetson AGX Xavier",32.369006156921,32.359191894531,32.367820978165
"Jetson Nano",150.4321539402,149.75622010231,150.3797390461
"Raspberry Pi 3 Model B+",1100.7618257999,1095.8513560295,1096.4616448879
"ASUS TinkerBoard",1527.2502539158,1414.4890110493,1407.0128879547,1968.4427921772,2033.5513451099,2575.5730659962
"ODROID-XU4",574.84182405472,573.67734313011,573.8012239933
"AGX Xavier 32.2",33.283438205719,33.393059015274,33.267495155334
"7-Zip Compression 16.02 - Compress Speed Test",
Higher Results Are Better
"Jetson TX1 Max-P",4529,4512,4483
"Jetson TX2 Max-Q",3313,3300,3269
"Jetson TX2 Max-P",5571,5635,5574
"Jetson AGX Xavier",16806,17652,19749,19582,19705,19527,19653,19780,19659,19415,19531,19483
"Jetson Nano",4084,4037,4025
"Raspberry Pi 3 Model B+",1778,2056,2047,2030,2041,2029,2031,2009,2039,2037,2042
"ASUS TinkerBoard",2866,2875,2766
"ODROID-XU4",4164,3810,4934,4463,4093,4087,3798,3949,4070,4053,3958,4064
"AGX Xavier 32.2",19659,19757,21448,21233,21142,21572,21187,21183,21224,21338,21241,21239
"Zstd Compression 1.3.4 - Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19",
Lower Results Are Better
"Jetson TX1 Max-P",146.09001684189,144.97263598442,146.3298740387
"Jetson TX2 Max-Q",252.57122015953,255.82780098915,252.99618387222
"Jetson TX2 Max-P",144.40146207809,145.3891749382,145.10904502869
"Jetson AGX Xavier",78.261996984482,81.176754951477,80.735954999924
"Jetson Nano",129.7424120903,130.30371499062,129.55208706856
"Raspberry Pi 3 Model B+",343.86739301682,342.51502299309,340.31952023506
"ASUS TinkerBoard",493.97650408745,500.89577293396,494.98889088631
"AGX Xavier 32.2",55.841320037842,56.245567083359,55.003093957901
"C-Ray 1.1 - Total Time - 4K, 16 Rays Per Pixel",
Lower Results Are Better
"Jetson TX1 Max-P",741.179,772.973,743.533
"Jetson TX2 Max-Q",865.785,870.174,870.017
"Jetson TX2 Max-P",977.761,533.547,534.592,538.351,536.314,533.971,548.331,531.981,532.463
"Jetson AGX Xavier",300.326,343.662,364.332,364.688,364.144,363.987,364.164,363.96,363.764
"Jetson Nano",921.883,920.709,921.024
"Raspberry Pi 3 Model B+",2034.332,2028.928,2025.911
"ASUS TinkerBoard",1673.941,1741.493,1738.825
"ODROID-XU4",1020.682,859.796,781.074,807.339,748.376,831.261,893.198,753.826,747.783
"AGX Xavier 32.2",164.472,162.347,162.004
"TTSIOD 3D Renderer 2.3b - Phong Rendering With Soft-Shadow Mapping",
Higher Results Are Better
"Jetson TX1 Max-P",45.0771,45.163,45.0377
"Jetson TX2 Max-Q",29.3872,29.3644,27.4862,29.1488
"Jetson TX2 Max-P",49.3978,49.4203,48.9575
"Jetson AGX Xavier",150.978,134.802,131.337,132.463,132.031,130.64,130.869,131.891,131.491,130.971,130.832,132.078
"Jetson Nano",40.9025,40.7681,41.1524
"Raspberry Pi 3 Model B+",17.984,17.5028,17.501
"ASUS TinkerBoard",23.3451,21.2351,20.9586,20.8803,20.8479,20.8715,20.9866,20.9135,20.9058
"ODROID-XU4",49.1282,42.8975,39.7692,40.4817,41.5918,42.5092,40.0077,41.0873,40.1642
"AGX Xavier 32.2",153.431,145.148,145.275,144.802
"OpenCV Benchmark 3.3.0 - ",
Lower Results Are Better
"Jetson TX2 Max-Q",504.22019696236,486.66019916534,487.34665894508
"Jetson TX2 Max-P",295.73621296883,295.94987106323,296.61644697189
"Jetson AGX Xavier",125.26104307175,130.6886138916,127.53062081337
"Jetson Nano",304.63343286514,263.82345294952,281.0886631012,271.52301716805,263.16651701927,264.02024912834,265.15014791489,264.3737039566,261.62518119812
"Raspberry Pi 3 Model B+",2.7371959686279
"ODROID-XU4",531.30143499374,514.98779511452,515.79874587059
"AGX Xavier 32.2",119.92672896385,119.71015501022,118.42247605324
"PyBench 2018-02-16 - Total For Average Test Times",
Lower Results Are Better
"Jetson TX1 Max-P",6351,6364,6303
"Jetson TX2 Max-Q",8690,8695,8820
"Jetson TX2 Max-P",5366,5475,5383
"Jetson AGX Xavier",2998,3008,3014
"Jetson Nano",7066,7156,7031
"Raspberry Pi 3 Model B+",20945,20826,20967
"ASUS TinkerBoard",15691,16297,10322,10167,10110,10339,10437,9624,10534
"ODROID-XU4",4990,5070,4968
"AGX Xavier 32.2",3009,2976,2949
"NVIDIA TensorRT Inference - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",39.9034,37.8756,39.6696
"Jetson TX2 Max-P",47.9151,50.0049,51.7866,50.166
"Jetson AGX Xavier",902.395,899.762,906.179
"Jetson Nano",21.5321,20.2945,21.068
"AGX Xavier 32.2",915.173,933.526,934.304
"NVIDIA TensorRT Inference - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",17.368,17.3589,17.3512
"Jetson TX2 Max-P",22.1156,22.049,22.031
"Jetson AGX Xavier",491.631,493.708,494.307
"AGX Xavier 32.2",511.725,512.508,512.336
"NVIDIA TensorRT Inference - Neural Network: AlexNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",235.623,235.707,239.839
"Jetson TX2 Max-P",301.864,300.93,300.061
"Jetson AGX Xavier",3142.28,3144.59,3140.95
"Jetson Nano",127.62,127.472,127.67
"AGX Xavier 32.2",3167.53,3167.53,3164.09
"NVIDIA TensorRT Inference - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",32.8097,32.4759,32.7174
"Jetson TX2 Max-P",42.0445,41.8598,41.8123
"Jetson AGX Xavier",260.15,260.002,259.311
"Jetson Nano",17.3879,17.3703,17.3957
"AGX Xavier 32.2",270.714,270.627,270.282
"NVIDIA TensorRT Inference - Neural Network: AlexNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",148.1,149.833,146.674
"Jetson TX2 Max-P",180.902,175.708,192.253,184.533,187.075
"Jetson AGX Xavier",1147.08,1138.12,1142.92
"Jetson Nano",82.8754,84.0409,85.382
"AGX Xavier 32.2",1138.28,1146.01,1130.21
"NVIDIA TensorRT Inference - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",103.681,103.755,103.513
"Jetson TX2 Max-P",131.414,128.848,130.083
"Jetson AGX Xavier",1700.28,1675.82,1703.42
"Jetson Nano",55.297,55.893,55.7755
"AGX Xavier 32.2",1740.15,1740.14,1741.27
"NVIDIA TensorRT Inference - Neural Network: VGG19 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",12.5342,12.6257,12.6031
"Jetson TX2 Max-P",15.9486,16.0019,15.8117
"Jetson AGX Xavier",394.465,395.111,394.399
"AGX Xavier 32.2",360.374,409.013,409.711,409.366,409.612,409.756,409.98,409.705,409.335,409.591,409.74,409.285
"NVIDIA TensorRT Inference - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",186.265,174.378,185.755,171.571,174.575,174.056,183.977,183.653
"Jetson TX2 Max-P",223.961,237.216,237.674
"Jetson AGX Xavier",1005.65,1006.32,1005.76
"Jetson Nano",99.0927,99.1556,98.5501
"AGX Xavier 32.2",1106.76,1161.7,1160.56
"NVIDIA TensorRT Inference - Neural Network: VGG16 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",15.81,15.7621,15.7914
"Jetson TX2 Max-P",19.8313,19.9879,19.9144
"Jetson AGX Xavier",475.195,474.877,475.154
"AGX Xavier 32.2",380.462,350.676,365.936,363.298,375.44,371.627
"NVIDIA TensorRT Inference - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",47.1592,47.2757,47.0066
"Jetson TX2 Max-P",59.7418,59.6222,59.7005
"Jetson AGX Xavier",1214.64,1215.5,1215.1
"Jetson Nano",25.0589,24.9801,25.1875
"AGX Xavier 32.2",1246.09,1248.2,1225.58
"NVIDIA TensorRT Inference - Neural Network: VGG19 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",11.6595,11.6931,10.9886
"Jetson TX2 Max-P",13.6304,14.3678,14.8003,14.462
"Jetson AGX Xavier",266.113,265.876,265.435
"AGX Xavier 32.2",271.323,271.525,272.103
"NVIDIA TensorRT Inference - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",84.3897,87.208,86.6317
"Jetson TX2 Max-P",110.76,112.798,108.572
"Jetson AGX Xavier",637.765,636.563,633.635
"Jetson Nano",46.5463,46.4845,46.4974
"AGX Xavier 32.2",699.466,699.937,700.272
"NVIDIA TensorRT Inference - Neural Network: VGG16 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",14.3354,14.5968,13.4569,14.3976,14.3904
"Jetson TX2 Max-P",16.3612,17.7045,17.707,17.77,17.727,18.1063
"Jetson AGX Xavier",303.711,303.005,304.611
"AGX Xavier 32.2",310.311,309.468,309.463
"NVIDIA TensorRT Inference - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",14.7989,14.4269,14.2882
"Jetson TX2 Max-P",18.4577,18.3856,18.0213
"Jetson AGX Xavier",375.594,372.502,370.096
"Jetson Nano",7.80364,7.71499,7.75225
"AGX Xavier 32.2",382.577,383.561,380.702
"CUDA Mini-Nbody 2015-11-10 - Test: Original",
Higher Results Are Better
"Jetson TX2 Max-Q",6.708,6.802,6.796
"Jetson TX2 Max-P",8.24,8.227,8.248
"Jetson AGX Xavier",47.134,47.124,47.135
"Jetson Nano",4.067,4.066,4.085
"AGX Xavier 32.2",46.76,42.767,40.898,37.79,36.145,35.696,34.774,33.317,32.275,32.091,32.022,31.835
"GLmark2 - Resolution: 1920 x 1080",
Higher Results Are Better
"Jetson AGX Xavier",
"Jetson Nano",
"AGX Xavier 32.2",
"TTSIOD 3D Renderer 2.3b - Performance / Cost - Phong Rendering With Soft-Shadow Mapping",
Higher Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"7-Zip Compression 16.02 - Performance / Cost - Compress Speed Test",
Higher Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"C-Ray 1.1 - Performance / Cost - Total Time - 4K, 16 Rays Per Pixel",
Lower Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"Rust Prime Benchmark - Performance / Cost - Prime Number Test To 200,000,000",
Lower Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"Zstd Compression 1.3.4 - Performance / Cost - Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19",
Lower Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"FLAC Audio Encoding 1.3.2 - Performance / Cost - WAV To FLAC",
Lower Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"PyBench 2018-02-16 - Performance / Cost - Total For Average Test Times",
Lower Results Are Better
"Jetson TX1 Max-P",
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ASUS TinkerBoard",
"ODROID-XU4",
"CUDA Mini-Nbody 2015-11-10 - Performance / Cost - Test: Original",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG16 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG16 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG19 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG19 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG16 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG16 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG19 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: VGG19 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: AlexNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: AlexNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: AlexNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: AlexNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 4 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet50 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet50 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: GoogleNet - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: GoogleNet - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet152 - Precision: FP16 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"NVIDIA TensorRT Inference - Performance / Cost - Neural Network: ResNet152 - Precision: INT8 - Batch Size: 32 - DLA Cores: Disabled",
Higher Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"OpenCV Benchmark 3.3.0 - Performance / Cost - ",
Lower Results Are Better
"Jetson TX2 Max-Q",
"Jetson TX2 Max-P",
"Jetson AGX Xavier",
"Jetson Nano",
"Raspberry Pi 3 Model B+",
"ODROID-XU4",
"GLmark2 - Performance / Cost - Resolution: 1920 x 1080",
Higher Results Are Better
"Jetson AGX Xavier",
"Jetson Nano",
"LeelaChessZero 0.20.1 - Performance / Cost - Backend: BLAS",
Higher Results Are Better
"Jetson AGX Xavier",
"Jetson Nano",
"LeelaChessZero 0.20.1 - Performance / Cost - Backend: CUDA + cuDNN",
Higher Results Are Better
"Jetson AGX Xavier",
"Jetson Nano",
"LeelaChessZero 0.20.1 - Performance / Cost - Backend: CUDA + cuDNN FP16",
Higher Results Are Better
"Jetson AGX Xavier",
"Tesseract OCR 4.0.0-beta.1 - Performance / Cost - Time To OCR 7 Images",
Lower Results Are Better
"Jetson AGX Xavier",
"Jetson Nano",
"ODROID-XU4",