Geekbench ML Score

2293
ONNX DirectML Inference Score
Geekbench ML 0.6.0 for Windows AArch64

Result Information

Upload Date July 18 2024 04:13 PM
Views 2

System Information

System Information
Operating System Microsoft Windows 11 Pro (64-bit)
Model LENOVO 21N10007UK
Motherboard LENOVO 21N10007UK
Power PlanBalanced
CPU Information
Name Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Oryon(TM) CPU
Topology 1 Processor, 12 Cores
Identifier ARMv8 (64-bit) Family 8 Model 1 Revision 201
Base Frequency 3.42 GHz
Cluster 18 Cores
Cluster 24 Cores
Memory Information
Size 32.00 GB
Inference Information
Framework ONNX
Backend DirectML
Device Snapdragon(R) X Elite - X1E78100 - Qualcomm(R) Adreno(TM) GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 1439
269.2 IPS
 
Image Classification (F16)
100% 1432
267.9 IPS
 
Image Classification (I8)
100% 1139
213.2 IPS
 
Image Segmentation (F32)
100% 1696
28.3 IPS
 
Image Segmentation (F16)
100% 1717
28.7 IPS
 
Image Segmentation (I8)
98% 1491
24.9 IPS
 
Pose Estimation (F32)
100% 9478
11.5 IPS
 
Pose Estimation (F16)
100% 9342
11.3 IPS
 
Pose Estimation (I8)
100% 7635
9.25 IPS
 
Object Detection (F32)
100% 1172
87.5 IPS
 
Object Detection (F16)
100% 1201
89.7 IPS
 
Object Detection (I8)
71% 1077
80.4 IPS
 
Face Detection (F32)
100% 2832
33.7 IPS
 
Face Detection (F16)
100% 2848
33.9 IPS
 
Face Detection (I8)
93% 2310
27.5 IPS
 
Depth Estimation (F32)
100% 5500
42.7 IPS
 
Depth Estimation (F16)
100% 5367
41.6 IPS
 
Depth Estimation (I8)
94% 4686
36.3 IPS
 
Style Transfer (F32)
100% 12936
17.0 IPS
 
Style Transfer (F16)
100% 13083
17.2 IPS
 
Style Transfer (I8)
98% 11824
15.6 IPS
 
Image Super-Resolution (F32)
100% 4839
172.8 IPS
 
Image Super-Resolution (F16)
100% 4919
175.7 IPS
 
Image Super-Resolution (I8)
99% 3581
127.9 IPS
 
Text Classification (F32)
100% 282
405.8 IPS
 
Text Classification (F16)
100% 285
409.1 IPS
 
Text Classification (I8)
98% 239
342.9 IPS
 
Machine Translation (F32)
100% 1055
19.4 IPS
 
Machine Translation (F16)
100% 1049
19.3 IPS
 
Machine Translation (I8)
65% 668
12.3 IPS
 

Compare

Set Baseline Geekbench 6 now available