Geekbench AI Score

190
Single Precision Score
191
Half Precision Score
616
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

User DoctorMurk
Upload Date October 05 2025 10:48 PM
Views 8

System Information

AI Information
Framework TensorFlow Lite
Backend QNN
Device ARM ARMv8
System Information
Operating System Android 15
Model samsung SM-X210
Model IDsamsung SM-X210
Motherboard holi
Governorschedutil
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 1
Base Frequency 1.80 GHz
Cluster 16 Cores @ 1.80 GHz
Cluster 22 Cores @ 2.21 GHz
Memory Information
Size 7.21 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 111
20.7 IPS
 
Image Classification (HP)
100% 113
20.9 IPS
 
Image Classification (Q)
96% 668
124.8 IPS
 
Image Segmentation (SP)
100% 152
2.47 IPS
 
Image Segmentation (HP)
100% 153
2.49 IPS
 
Image Segmentation (Q)
98% 864
14.1 IPS
 
Pose Estimation (SP)
100% 195
0.23 IPS
 
Pose Estimation (HP)
100% 198
0.23 IPS
 
Pose Estimation (Q)
96% 5120
6.00 IPS
 
Object Detection (SP)
100% 109
8.65 IPS
 
Object Detection (HP)
100% 109
8.66 IPS
 
Object Detection (Q)
84% 802
64.7 IPS
 
Face Detection (SP)
100% 286
3.40 IPS
 
Face Detection (HP)
100% 286
3.40 IPS
 
Face Detection (Q)
97% 1468
17.5 IPS
 
Depth Estimation (SP)
100% 245
1.88 IPS
 
Depth Estimation (HP)
99% 245
1.89 IPS
 
Depth Estimation (Q)
64% 3006
27.6 IPS
 
Style Transfer (SP)
100% 427
0.55 IPS
 
Style Transfer (HP)
100% 430
0.55 IPS
 
Style Transfer (Q)
98% 8323
10.7 IPS
 
Image Super-Resolution (SP)
100% 125
4.61 IPS
 
Image Super-Resolution (HP)
100% 125
4.63 IPS
 
Image Super-Resolution (Q)
97% 433
16.1 IPS
 
Text Classification (SP)
100% 165
220.5 IPS
 
Text Classification (HP)
100% 166
221.0 IPS
 
Text Classification (Q)
93% 212
284.6 IPS
 
Machine Translation (SP)
100% 282
4.85 IPS
 
Machine Translation (HP)
100% 283
4.87 IPS
 
Machine Translation (Q)
0% 0
1.67 IPS
 

Compare

Set Baseline Geekbench 6 now available