Geekbench AI Score

171
Single Precision Score
191
Half Precision Score
472
Quantized Score
Geekbench AI 1.4.0 for Android AArch64

Result Information

User M91
Upload Date August 31 2025 08:41 PM
Views 5

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 15
Model samsung SM-A166W
Model IDsamsung SM-A166W
Motherboard s5e8535
Governorenergy_aware
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 1
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.40 GHz
Memory Information
Size 3.37 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 122
22.7 IPS
 
Image Classification (HP)
100% 122
22.8 IPS
 
Image Classification (Q)
99% 328
61.1 IPS
 
Image Segmentation (SP)
100% 168
2.72 IPS
 
Image Segmentation (HP)
100% 169
2.74 IPS
 
Image Segmentation (Q)
98% 372
6.06 IPS
 
Pose Estimation (SP)
100% 220
0.26 IPS
 
Pose Estimation (HP)
100% 217
0.25 IPS
 
Pose Estimation (Q)
98% 867
1.02 IPS
 
Object Detection (SP)
100% 123
9.73 IPS
 
Object Detection (HP)
100% 123
9.79 IPS
 
Object Detection (Q)
87% 352
28.2 IPS
 
Face Detection (SP)
100% 317
3.77 IPS
 
Face Detection (HP)
100% 315
3.74 IPS
 
Face Detection (Q)
97% 710
8.47 IPS
 
Depth Estimation (SP)
100% 270
2.08 IPS
 
Depth Estimation (HP)
99% 260
2.00 IPS
 
Depth Estimation (Q)
64% 638
5.79 IPS
 
Style Transfer (SP)
100% 98
0.13 IPS
 
Style Transfer (HP)
100% 291
0.37 IPS
 
Style Transfer (Q)
98% 1033
1.33 IPS
 
Image Super-Resolution (SP)
100% 113
4.16 IPS
 
Image Super-Resolution (HP)
100% 117
4.34 IPS
 
Image Super-Resolution (Q)
97% 405
15.0 IPS
 
Text Classification (SP)
100% 157
209.8 IPS
 
Text Classification (HP)
100% 157
209.8 IPS
 
Text Classification (Q)
91% 299
401.7 IPS
 
Machine Translation (SP)
100% 271
4.67 IPS
 
Machine Translation (HP)
100% 270
4.65 IPS
 
Machine Translation (Q)
57% 262
6.40 IPS
 

Compare

Set Baseline Geekbench 6 now available