Geekbench AI Score

529
Single Precision Score
579
Half Precision Score
1693
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

User inc7007
Upload Date June 13 2025 10:21 AM
Views 3

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 15
Model samsung SM-A346B
Model IDsamsung SM-A346B
Motherboard a34x
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 0
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.60 GHz
Memory Information
Size 7.30 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 374
69.5 IPS
 
Image Classification (HP)
100% 1254
233.1 IPS
 
Image Classification (Q)
99% 1745
325.4 IPS
 
Image Segmentation (SP)
100% 488
7.91 IPS
 
Image Segmentation (HP)
100% 99
1.61 IPS
 
Image Segmentation (Q)
98% 1778
28.9 IPS
 
Pose Estimation (SP)
100% 2350
2.74 IPS
 
Pose Estimation (HP)
100% 4312
5.03 IPS
 
Pose Estimation (Q)
95% 10707
12.6 IPS
 
Object Detection (SP)
100% 288
22.9 IPS
 
Object Detection (HP)
100% 1215
96.4 IPS
 
Object Detection (Q)
89% 1576
126.2 IPS
 
Face Detection (SP)
100% 1188
14.1 IPS
 
Face Detection (HP)
100% 2938
34.9 IPS
 
Face Detection (Q)
97% 5318
63.4 IPS
 
Depth Estimation (SP)
100% 385
2.97 IPS
 
Depth Estimation (HP)
99% 157
1.21 IPS
 
Depth Estimation (Q)
64% 1321
12.1 IPS
 
Style Transfer (SP)
100% 1573
2.02 IPS
 
Style Transfer (HP)
100% 284
0.37 IPS
 
Style Transfer (Q)
98% 2858
3.69 IPS
 
Image Super-Resolution (SP)
100% 896
33.1 IPS
 
Image Super-Resolution (HP)
100% 2907
107.4 IPS
 
Image Super-Resolution (Q)
66% 3898
165.8 IPS
 
Text Classification (SP)
100% 117
156.5 IPS
 
Text Classification (HP)
100% 102
135.6 IPS
 
Text Classification (Q)
93% 291
391.3 IPS
 
Machine Translation (SP)
100% 186
3.20 IPS
 
Machine Translation (HP)
100% 168
2.89 IPS
 
Machine Translation (Q)
51% 163
5.39 IPS
 

Compare

Set Baseline Geekbench 6 now available