Geekbench AI Score

100
Single Precision Score
157
Half Precision Score
154
Quantized Score
Geekbench AI 1.0.0 for Android AArch64

Result Information

Upload Date August 22 2024 09:51 PM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend GPU
Device Qualcomm Snapdragon 670
System Information
Operating System Android 14
Model Samsung Galaxy Tab S5e
Model IDsamsung SM-T720
Motherboard sdm710
Governorschedutil
CPU Information
Name Qualcomm Qualcomm
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 6 part 2050 revision 13
Base Frequency 1.71 GHz
Cluster 16 Cores @ 1.71 GHz
Cluster 22 Cores @ 2.00 GHz
Memory Information
Size 3.56 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 61
11.3 IPS
 
Image Classification (HP)
100% 103
19.1 IPS
 
Image Classification (Q)
97% 107
20.0 IPS
 
Image Segmentation (SP)
100% 71
1.15 IPS
 
Image Segmentation (HP)
100% 115
1.87 IPS
 
Image Segmentation (Q)
98% 125
2.03 IPS
 
Pose Estimation (SP)
100% 117
0.14 IPS
 
Pose Estimation (HP)
100% 185
0.22 IPS
 
Pose Estimation (Q)
95% 184
0.22 IPS
 
Object Detection (SP)
100% 53
4.24 IPS
 
Object Detection (HP)
100% 84
6.69 IPS
 
Object Detection (Q)
86% 87
7.00 IPS
 
Face Detection (SP)
100% 144
1.71 IPS
 
Face Detection (HP)
100% 259
3.08 IPS
 
Face Detection (Q)
97% 272
3.24 IPS
 
Depth Estimation (SP)
100% 145
1.12 IPS
 
Depth Estimation (HP)
99% 266
2.05 IPS
 
Depth Estimation (Q)
75% 255
2.06 IPS
 
Style Transfer (SP)
100% 277
0.36 IPS
 
Style Transfer (HP)
100% 634
0.82 IPS
 
Style Transfer (Q)
98% 631
0.81 IPS
 
Image Super-Resolution (SP)
100% 71
2.61 IPS
 
Image Super-Resolution (HP)
100% 128
4.72 IPS
 
Image Super-Resolution (Q)
97% 126
4.68 IPS
 
Text Classification (SP)
100% 73
96.9 IPS
 
Text Classification (HP)
100% 73
97.0 IPS
 
Text Classification (Q)
91% 84
112.5 IPS
 
Machine Translation (SP)
100% 129
2.23 IPS
 
Machine Translation (HP)
100% 129
2.22 IPS
 
Machine Translation (Q)
57% 77
1.87 IPS
 

Compare

Set Baseline Geekbench 6 now available