Geekbench AI Score

488
Single Precision Score
168
Half Precision Score
172
Quantized Score
Geekbench AI 1.0.0 for Windows AArch64

Result Information

User grinceur
Upload Date August 18 2024 04:53 PM
Views 8

System Information

AI Information
Framework ONNX
Backend QNN
Device Microsoft SQ1 @ 3.0 GHz
System Information
Operating System Microsoft Windows 11 Professionnel (64-bit)
Model Microsoft Corporation Surface Pro X
Motherboard Microsoft Corporation Surface Pro X
Power PlanUtilisation normale
CPU Information
Name Microsoft SQ1 @ 3.0 GHz
Topology 1 Processor, 8 Cores
Identifier ARMv8 (64-bit) Family 8 Model 805 Revision D0E
Base Frequency 3.00 GHz
Cluster 14 Cores
Cluster 24 Cores
Memory Information
Size 8.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 214
39.8 IPS
 
Image Classification (HP)
100% 64
11.9 IPS
 
Image Classification (Q)
100% 105
19.5 IPS
 
Image Segmentation (SP)
100% 258
4.18 IPS
 
Image Segmentation (HP)
100% 97
1.58 IPS
 
Image Segmentation (Q)
98% 122
1.98 IPS
 
Pose Estimation (SP)
100% 840
0.98 IPS
 
Pose Estimation (HP)
100% 290
0.34 IPS
 
Pose Estimation (Q)
95% 247
0.29 IPS
 
Object Detection (SP)
100% 217
17.2 IPS
 
Object Detection (HP)
99% 71
5.64 IPS
 
Object Detection (Q)
89% 90
7.23 IPS
 
Face Detection (SP)
100% 626
7.44 IPS
 
Face Detection (HP)
100% 126
1.49 IPS
 
Face Detection (Q)
97% 244
2.91 IPS
 
Depth Estimation (SP)
100% 638
4.92 IPS
 
Depth Estimation (HP)
99% 264
2.04 IPS
 
Depth Estimation (Q)
73% 257
2.10 IPS
 
Style Transfer (SP)
100% 2174
2.79 IPS
 
Style Transfer (HP)
100% 1063
1.37 IPS
 
Style Transfer (Q)
98% 647
0.83 IPS
 
Image Super-Resolution (SP)
100% 405
15.0 IPS
 
Image Super-Resolution (HP)
100% 187
6.89 IPS
 
Image Super-Resolution (Q)
99% 140
5.17 IPS
 
Text Classification (SP)
100% 437
583.8 IPS
 
Text Classification (HP)
100% 124
166.2 IPS
 
Text Classification (Q)
97% 106
142.3 IPS
 
Machine Translation (SP)
100% 499
8.59 IPS
 
Machine Translation (HP)
99% 174
3.00 IPS
 
Machine Translation (Q)
59% 138
3.20 IPS
 

Compare

Set Baseline Geekbench 6 now available