ARMv8 Neoverse-V2 testing with a Quanta Cloud QuantaGrid S74G-2U 1S7GZ9Z0000 S7G MB (CG1) (3A06 BIOS) and ASPEED on Ubuntu 23.10 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 2402042-NE-ONNXGH20009
onnx gh200,
"ONNX Runtime 1.17 - Model: GPT-2 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: GPT-2 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: yolov4 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",8.41599,8.32484,8.35707
"b",8.83004,8.06401,8.19663,8.40329,8.23359,8.36832,8.23278,7.988,8.31405,8.34558,8.39933,8.31301,8.318
"c",
"d",
"ONNX Runtime 1.17 - Model: yolov4 - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",118.818,120.118,119.655
"b",113.246,124.004,121.998,118.997,121.451,119.495,121.462,125.184,120.274,119.82,119.053,120.29,120.218
"c",
"d",
"ONNX Runtime 1.17 - Model: yolov4 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",11.2301,10.8986,11.0773
"b",11.2771,11.2389,11.0394
"c",
"d",
"ONNX Runtime 1.17 - Model: yolov4 - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",89.0387,91.7461,90.2656
"b",88.6674,88.9678,90.5772
"c",
"d",
"ONNX Runtime 1.17 - Model: T5 Encoder - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",74.598,74.6921,76.4351
"b",73.6533,75.3117,73.5266
"c",
"d",
"ONNX Runtime 1.17 - Model: T5 Encoder - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",13.4026,13.3835,13.0807
"b",13.574,13.275,13.5976
"c",
"d",
"ONNX Runtime 1.17 - Model: T5 Encoder - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",341.859,368.568,365.37,382.276,353.875,380.891,365.558,362.644,390.943,333.792,357.906,376.676,387.59,375.825,375.674
"b",363.986,339.27,331.162,371.535,378.943,380.673,391.079,393.718,392.798,387.834,375.002,364.73,385.388,384.26,372.726
"c",
"d",
"ONNX Runtime 1.17 - Model: T5 Encoder - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",2.91737,2.70531,2.7312,2.60846,2.81871,2.62077,2.72992,2.75164,2.55277,2.98775,2.78733,2.64892,2.57464,2.65635,2.65633
"b",2.74029,2.93992,3.01203,2.68472,2.63263,2.62222,2.55182,2.53541,2.54007,2.57321,2.66109,2.73732,2.5905,2.59725,2.67748
"c",
"d",
"ONNX Runtime 1.17 - Model: bertsquad-12 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: bertsquad-12 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",552.668,553.045,541.183
"b",533.381,536.615,526.788
"c",
"d",
"ONNX Runtime 1.17 - Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",1.80772,1.80643,1.84611
"b",1.87304,1.86172,1.89652
"c",
"d",
"ONNX Runtime 1.17 - Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",1123.28,1101.64,1113.15
"b",1146.1,1113.86,1116.68
"c",
"d",
"ONNX Runtime 1.17 - Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",0.889227,0.906567,0.897466
"b",0.871628,0.896721,0.894516
"c",
"d",
"ONNX Runtime 1.17 - Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",1.50841,1.41845,1.49666,1.43536,1.50079,1.4743,1.49948
"b",1.47791,1.49046,1.46398
"c",
"d",
"ONNX Runtime 1.17 - Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",662.948,704.986,668.149,696.686,666.312,678.286,666.892
"b",676.626,670.929,683.067
"c",
"d",
"ONNX Runtime 1.17 - Model: fcn-resnet101-11 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",1.47684,1.48748,1.49024
"b",1.51187,1.49066,1.48121
"c",
"d",
"ONNX Runtime 1.17 - Model: fcn-resnet101-11 - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",677.115,672.269,671.025
"b",661.425,670.835,675.116
"c",
"d",
"ONNX Runtime 1.17 - Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",243.02,243.608,237.656
"b",244.614,240.363,238.326
"c",
"d",
"ONNX Runtime 1.17 - Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",4.11359,4.10362,4.20636
"b",4.08673,4.15899,4.19431
"c",
"d",
"ONNX Runtime 1.17 - Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",268.788,278.228,271.027
"b",283.441,273.83,275.212
"c",
"d",
"ONNX Runtime 1.17 - Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",3.71904,3.59274,3.68826
"b",3.52659,3.65053,3.63193
"c",
"d",
"ONNX Runtime 1.17 - Model: super-resolution-10 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",43.2509,44.1195,42.7406
"b",43.5197,42.4651,42.3369
"c",
"d",
"ONNX Runtime 1.17 - Model: super-resolution-10 - Device: CPU - Executor: Parallel",
Lower Results Are Better
"a",23.1201,22.6649,23.3962
"b",22.9772,23.548,23.6193
"c",
"d",
"ONNX Runtime 1.17 - Model: super-resolution-10 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",149.797,149.219,149.619
"b",149.701,148.704,147.738
"c",
"d",
"ONNX Runtime 1.17 - Model: super-resolution-10 - Device: CPU - Executor: Standard",
Lower Results Are Better
"a",6.674,6.69952,6.68265
"b",6.67717,6.7229,6.76653
"c",
"d",
"ONNX Runtime 1.17 - Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel",
Higher Results Are Better
"a",
"b",
"c",
"d",
"ONNX Runtime 1.17 - Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard",
Higher Results Are Better
"a",
"b",
"c",
"d",