Geekbench ML Score

310
TensorFlow Lite GPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date September 02 2024 09:27 AM
Views 1

System Information

System Information
Operating System Android 14
Model Samsung Galaxy A52S 5G
Model IDsamsung SM-A528B
Motherboard lahaina
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 14 Cores @ 1.80 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 5.25 GB
Inference Information
Framework TensorFlow Lite
Backend GPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 166
31.0 IPS
 
Image Classification (F16)
100% 214
40.1 IPS
 
Image Classification (I8)
97% 209
39.0 IPS
 
Image Segmentation (F32)
100% 277
4.63 IPS
 
Image Segmentation (F16)
100% 366
6.10 IPS
 
Image Segmentation (I8)
98% 377
6.30 IPS
 
Pose Estimation (F32)
100% 255
0.31 IPS
 
Pose Estimation (F16)
100% 180
0.22 IPS
 
Pose Estimation (I8)
100% 179
0.22 IPS
 
Object Detection (F32)
100% 135
10.1 IPS
 
Object Detection (F16)
100% 159
11.8 IPS
 
Object Detection (I8)
62% 154
11.5 IPS
 
Face Detection (F32)
100% 563
6.70 IPS
 
Face Detection (F16)
97% 810
9.63 IPS
 
Face Detection (I8)
87% 781
9.29 IPS
 
Depth Estimation (F32)
100% 406
3.15 IPS
 
Depth Estimation (F16)
100% 610
4.73 IPS
 
Depth Estimation (I8)
95% 609
4.72 IPS
 
Style Transfer (F32)
100% 781
1.03 IPS
 
Style Transfer (F16)
100% 1332
1.75 IPS
 
Style Transfer (I8)
98% 1328
1.75 IPS
 
Image Super-Resolution (F32)
100% 181
6.45 IPS
 
Image Super-Resolution (F16)
100% 320
11.4 IPS
 
Image Super-Resolution (I8)
98% 311
11.1 IPS
 
Text Classification (F32)
100% 125
179.9 IPS
 
Text Classification (F16)
100% 125
179.9 IPS
 
Text Classification (I8)
92% 233
334.4 IPS
 
Machine Translation (F32)
100% 231
4.24 IPS
 
Machine Translation (F16)
100% 230
4.22 IPS
 
Machine Translation (I8)
64% 293
5.39 IPS
 

Compare

Set Baseline Geekbench 6 now available