Geekbench AI Score

698
Single Precision Score
696
Half Precision Score
1432
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

Upload Date May 06 2025 02:18 AM
Views 3

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARM ARMv8
System Information
Operating System Android 14
Model samsung SM-A356N
Model IDsamsung SM-A356N
Motherboard s5e8835
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 14 Cores @ 2.00 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 5.30 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 388
72.2 IPS
 
Image Classification (HP)
100% 389
72.3 IPS
 
Image Classification (Q)
100% 1267
235.6 IPS
 
Image Segmentation (SP)
100% 466
7.55 IPS
 
Image Segmentation (HP)
100% 481
7.80 IPS
 
Image Segmentation (Q)
98% 1428
23.2 IPS
 
Pose Estimation (SP)
100% 1588
1.85 IPS
 
Pose Estimation (HP)
100% 1496
1.75 IPS
 
Pose Estimation (Q)
84% 2631
3.12 IPS
 
Object Detection (SP)
100% 313
24.8 IPS
 
Object Detection (HP)
100% 310
24.6 IPS
 
Object Detection (Q)
83% 958
77.4 IPS
 
Face Detection (SP)
100% 994
11.8 IPS
 
Face Detection (HP)
100% 998
11.9 IPS
 
Face Detection (Q)
95% 2354
28.1 IPS
 
Depth Estimation (SP)
100% 1211
9.33 IPS
 
Depth Estimation (HP)
99% 1257
9.69 IPS
 
Depth Estimation (Q)
64% 2255
20.7 IPS
 
Style Transfer (SP)
89% 2502
3.25 IPS
 
Style Transfer (HP)
89% 2518
3.27 IPS
 
Style Transfer (Q)
98% 5721
7.38 IPS
 
Image Super-Resolution (SP)
100% 645
23.8 IPS
 
Image Super-Resolution (HP)
100% 643
23.7 IPS
 
Image Super-Resolution (Q)
97% 1355
50.2 IPS
 
Text Classification (SP)
100% 320
427.0 IPS
 
Text Classification (HP)
100% 318
424.4 IPS
 
Text Classification (Q)
88% 606
817.2 IPS
 
Machine Translation (SP)
100% 492
8.48 IPS
 
Machine Translation (HP)
100% 477
8.21 IPS
 
Machine Translation (Q)
50% 319
11.3 IPS
 

Compare

Set Baseline Geekbench 6 now available