Geekbench AI Score

534
Single Precision Score
939
Half Precision Score
747
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

User M91
Upload Date October 22 2025 11:08 AM
Views 2

System Information

AI Information
Framework TensorFlow Lite
Backend GPU
Device ARM ARMv8
System Information
Operating System Android 16
Model samsung SM-A366W
Model IDsamsung SM-A366W
Motherboard parrot
Governorwalt
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 2
Base Frequency 1.80 GHz
Cluster 14 Cores @ 1.80 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 5.25 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 410
76.2 IPS
 
Image Classification (HP)
100% 686
127.6 IPS
 
Image Classification (Q)
100% 647
120.4 IPS
 
Image Segmentation (SP)
100% 583
9.45 IPS
 
Image Segmentation (HP)
100% 1034
16.8 IPS
 
Image Segmentation (Q)
98% 979
15.9 IPS
 
Pose Estimation (SP)
100% 2005
2.34 IPS
 
Pose Estimation (HP)
99% 3841
4.49 IPS
 
Pose Estimation (Q)
96% 3799
4.45 IPS
 
Object Detection (SP)
100% 352
28.0 IPS
 
Object Detection (HP)
99% 580
46.0 IPS
 
Object Detection (Q)
87% 554
44.5 IPS
 
Face Detection (SP)
100% 1041
12.4 IPS
 
Face Detection (HP)
100% 2049
24.4 IPS
 
Face Detection (Q)
97% 1783
21.3 IPS
 
Depth Estimation (SP)
100% 1179
9.08 IPS
 
Depth Estimation (HP)
98% 2310
17.9 IPS
 
Depth Estimation (Q)
61% 1716
16.9 IPS
 
Style Transfer (SP)
100% 2170
2.79 IPS
 
Style Transfer (HP)
100% 4555
5.86 IPS
 
Style Transfer (Q)
98% 4513
5.82 IPS
 
Image Super-Resolution (SP)
100% 506
18.7 IPS
 
Image Super-Resolution (HP)
100% 1037
38.3 IPS
 
Image Super-Resolution (Q)
97% 1020
37.8 IPS
 
Text Classification (SP)
35% 22
217.5 IPS
 
Text Classification (HP)
35% 35
314.4 IPS
 
Text Classification (Q)
33% 26
311.5 IPS
 
Machine Translation (SP)
100% 378
6.50 IPS
 
Machine Translation (HP)
97% 433
7.49 IPS
 
Machine Translation (Q)
43% 111
6.30 IPS
 

Compare

Set Baseline Geekbench 6 now available