Geekbench AI Score

182
Single Precision Score
182
Half Precision Score
441
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

Upload Date June 14 2025 08:01 PM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device Samsung Exynos 2200
System Information
Operating System Android 15
Model Samsung Galaxy S22+
Model IDsamsung SM-S906B
Motherboard s5e9925
Governorenergy_aware
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 2 part 3400 revision 0
Base Frequency 1.82 GHz
Cluster 14 Cores @ 1.82 GHz
Cluster 23 Cores @ 2.52 GHz
Cluster 31 Core @ 2.80 GHz
Memory Information
Size 7.10 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 105
19.5 IPS
 
Image Classification (HP)
100% 105
19.6 IPS
 
Image Classification (Q)
99% 296
55.2 IPS
 
Image Segmentation (SP)
100% 138
2.24 IPS
 
Image Segmentation (HP)
100% 138
2.24 IPS
 
Image Segmentation (Q)
98% 315
5.13 IPS
 
Pose Estimation (SP)
100% 190
0.22 IPS
 
Pose Estimation (HP)
100% 188
0.22 IPS
 
Pose Estimation (Q)
98% 752
0.88 IPS
 
Object Detection (SP)
100% 104
8.22 IPS
 
Object Detection (HP)
100% 104
8.23 IPS
 
Object Detection (Q)
87% 319
25.6 IPS
 
Face Detection (SP)
100% 296
3.51 IPS
 
Face Detection (HP)
100% 297
3.53 IPS
 
Face Detection (Q)
97% 687
8.20 IPS
 
Depth Estimation (SP)
100% 245
1.88 IPS
 
Depth Estimation (HP)
99% 248
1.91 IPS
 
Depth Estimation (Q)
64% 609
5.53 IPS
 
Style Transfer (SP)
100% 439
0.56 IPS
 
Style Transfer (HP)
100% 436
0.56 IPS
 
Style Transfer (Q)
98% 1112
1.43 IPS
 
Image Super-Resolution (SP)
100% 118
4.35 IPS
 
Image Super-Resolution (HP)
100% 117
4.32 IPS
 
Image Super-Resolution (Q)
97% 434
16.1 IPS
 
Text Classification (SP)
100% 147
195.6 IPS
 
Text Classification (HP)
100% 146
195.1 IPS
 
Text Classification (Q)
91% 269
361.8 IPS
 
Machine Translation (SP)
100% 260
4.48 IPS
 
Machine Translation (HP)
100% 260
4.48 IPS
 
Machine Translation (Q)
57% 232
5.66 IPS
 

Compare

Set Baseline Geekbench 6 now available