Geekbench AI Score

1910
Single Precision Score
1881
Half Precision Score
3087
Quantized Score
Geekbench AI 1.6.1 for Android AArch64

Result Information

Upload Date February 10 2026 05:21 PM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device Qualcomm Snapdragon 8 Elite
System Information
Operating System Android 16
Model Samsung Galaxy S25
Model IDsamsung SM-S931B
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% 1311
243.8 IPS
 
Image Classification (HP)
100% 1589
295.5 IPS
 
Image Classification (Q)
100% 3503
651.4 IPS
 
Image Segmentation (SP)
100% 1821
29.5 IPS
 
Image Segmentation (HP)
100% 1829
29.7 IPS
 
Image Segmentation (Q)
98% 4065
66.1 IPS
 
Pose Estimation (SP)
100% 4192
4.89 IPS
 
Pose Estimation (HP)
100% 2988
3.49 IPS
 
Pose Estimation (Q)
84% 8303
9.85 IPS
 
Object Detection (SP)
98% 969
77.1 IPS
 
Object Detection (HP)
98% 980
78.0 IPS
 
Object Detection (Q)
83% 2551
206.0 IPS
 
Face Detection (SP)
100% 3131
37.2 IPS
 
Face Detection (HP)
100% 2966
35.2 IPS
 
Face Detection (Q)
95% 6544
78.1 IPS
 
Depth Estimation (SP)
99% 3120
24.1 IPS
 
Depth Estimation (HP)
99% 2977
23.0 IPS
 
Depth Estimation (Q)
64% 6361
58.3 IPS
 
Style Transfer (SP)
89% 8614
11.2 IPS
 
Style Transfer (HP)
89% 6984
9.07 IPS
 
Style Transfer (Q)
98% 13764
17.7 IPS
 
Image Super-Resolution (SP)
100% 1382
51.0 IPS
 
Image Super-Resolution (HP)
100% 1548
57.2 IPS
 
Image Super-Resolution (Q)
97% 3510
130.1 IPS
 
Text Classification (SP)
100% 460
613.5 IPS
 
Text Classification (HP)
99% 644
860.2 IPS
 
Text Classification (Q)
88% 398
537.5 IPS
 
Machine Translation (SP)
100% 1248
21.5 IPS
 
Machine Translation (HP)
100% 1062
18.3 IPS
 
Machine Translation (Q)
50% 326
11.6 IPS
 

Compare

Set Baseline Geekbench 6 now available