Geekbench AI Score

545
Single Precision Score
552
Half Precision Score
1171
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

Upload Date November 07 2025 09:27 AM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device Qualcomm Snapdragon 8 Elite
System Information
Operating System Android 16
Model Samsung Galaxy S25 Ultra
Model IDsamsung SM-S938N
Motherboard sun
Governorwalt
CPU Information
Name Qualcomm ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 3 part 1 revision 4
Base Frequency 3.53 GHz
Cluster 16 Cores @ 3.53 GHz
Cluster 22 Cores @ 4.47 GHz
Memory Information
Size 10.85 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 369
68.6 IPS
 
Image Classification (HP)
100% 375
69.8 IPS
 
Image Classification (Q)
99% 929
173.2 IPS
 
Image Segmentation (SP)
100% 490
7.95 IPS
 
Image Segmentation (HP)
100% 463
7.50 IPS
 
Image Segmentation (Q)
98% 878
14.3 IPS
 
Pose Estimation (SP)
100% 607
0.71 IPS
 
Pose Estimation (HP)
100% 596
0.70 IPS
 
Pose Estimation (Q)
98% 2089
2.45 IPS
 
Object Detection (SP)
100% 303
24.0 IPS
 
Object Detection (HP)
100% 333
26.4 IPS
 
Object Detection (Q)
87% 891
71.5 IPS
 
Face Detection (SP)
100% 848
10.1 IPS
 
Face Detection (HP)
100% 778
9.25 IPS
 
Face Detection (Q)
97% 1802
21.5 IPS
 
Depth Estimation (SP)
100% 682
5.25 IPS
 
Depth Estimation (HP)
99% 755
5.82 IPS
 
Depth Estimation (Q)
64% 1523
13.8 IPS
 
Style Transfer (SP)
100% 1169
1.50 IPS
 
Style Transfer (HP)
100% 1085
1.39 IPS
 
Style Transfer (Q)
98% 2891
3.73 IPS
 
Image Super-Resolution (SP)
100% 309
11.4 IPS
 
Image Super-Resolution (HP)
100% 338
12.5 IPS
 
Image Super-Resolution (Q)
97% 1204
44.6 IPS
 
Text Classification (SP)
100% 433
578.4 IPS
 
Text Classification (HP)
100% 467
622.8 IPS
 
Text Classification (Q)
91% 718
964.6 IPS
 
Machine Translation (SP)
100% 769
13.2 IPS
 
Machine Translation (HP)
100% 767
13.2 IPS
 
Machine Translation (Q)
57% 467
11.4 IPS
 

Compare

Set Baseline Geekbench 6 now available