Geekbench AI Score

264
Single Precision Score
262
Half Precision Score
280
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

User eunu
Upload Date October 08 2025 09:19 AM
Views 8

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARM Qualcomm
System Information
Operating System Android 14
Model samsung SM-A235N
Model IDsamsung SM-A235N
Motherboard bengal
Governorschedutil
CPU Information
Name ARM Qualcomm
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3337 revision 0
Base Frequency 1.90 GHz
Cluster 14 Cores @ 1.90 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 3.50 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 131
24.3 IPS
 
Image Classification (HP)
100% 132
24.6 IPS
 
Image Classification (Q)
100% 231
43.0 IPS
 
Image Segmentation (SP)
100% 168
2.73 IPS
 
Image Segmentation (HP)
100% 168
2.72 IPS
 
Image Segmentation (Q)
98% 297
4.83 IPS
 
Pose Estimation (SP)
100% 655
0.76 IPS
 
Pose Estimation (HP)
100% 655
0.76 IPS
 
Pose Estimation (Q)
84% 507
0.60 IPS
 
Object Detection (SP)
100% 129
10.3 IPS
 
Object Detection (HP)
100% 123
9.77 IPS
 
Object Detection (Q)
83% 231
18.7 IPS
 
Face Detection (SP)
100% 400
4.76 IPS
 
Face Detection (HP)
100% 397
4.72 IPS
 
Face Detection (Q)
95% 557
6.65 IPS
 
Depth Estimation (SP)
100% 562
4.33 IPS
 
Depth Estimation (HP)
99% 559
4.30 IPS
 
Depth Estimation (Q)
64% 336
3.08 IPS
 
Style Transfer (SP)
89% 852
1.11 IPS
 
Style Transfer (HP)
89% 853
1.11 IPS
 
Style Transfer (Q)
98% 847
1.09 IPS
 
Image Super-Resolution (SP)
100% 198
7.30 IPS
 
Image Super-Resolution (HP)
100% 196
7.23 IPS
 
Image Super-Resolution (Q)
97% 197
7.30 IPS
 
Text Classification (SP)
100% 121
161.2 IPS
 
Text Classification (HP)
100% 119
158.8 IPS
 
Text Classification (Q)
88% 132
178.3 IPS
 
Machine Translation (SP)
100% 198
3.41 IPS
 
Machine Translation (HP)
100% 198
3.42 IPS
 
Machine Translation (Q)
50% 92
3.28 IPS
 

Compare

Set Baseline Geekbench 6 now available