Geekbench ML Score

560
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date July 04 2024 07:38 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 7.19 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 261
48.8 IPS
 
Image Classification (F16)
100% 263
49.2 IPS
 
Image Classification (I8)
97% 643
120.3 IPS
 
Image Segmentation (F32)
100% 346
5.77 IPS
 
Image Segmentation (F16)
100% 343
5.74 IPS
 
Image Segmentation (I8)
98% 508
8.48 IPS
 
Pose Estimation (F32)
100% 520
0.63 IPS
 
Pose Estimation (F16)
100% 512
0.62 IPS
 
Pose Estimation (I8)
100% 2619
3.17 IPS
 
Object Detection (F32)
100% 283
21.2 IPS
 
Object Detection (F16)
100% 286
21.4 IPS
 
Object Detection (I8)
61% 752
56.1 IPS
 
Face Detection (F32)
100% 531
6.31 IPS
 
Face Detection (F16)
100% 530
6.30 IPS
 
Face Detection (I8)
86% 1466
17.4 IPS
 
Depth Estimation (F32)
100% 535
4.15 IPS
 
Depth Estimation (F16)
100% 531
4.12 IPS
 
Depth Estimation (I8)
95% 1419
11.0 IPS
 
Style Transfer (F32)
100% 930
1.22 IPS
 
Style Transfer (F16)
100% 926
1.22 IPS
 
Style Transfer (I8)
98% 2544
3.35 IPS
 
Image Super-Resolution (F32)
100% 356
12.7 IPS
 
Image Super-Resolution (F16)
100% 365
13.0 IPS
 
Image Super-Resolution (I8)
98% 1243
44.4 IPS
 
Text Classification (F32)
100% 354
509.1 IPS
 
Text Classification (F16)
100% 344
494.4 IPS
 
Text Classification (I8)
92% 316
454.6 IPS
 
Machine Translation (F32)
100% 626
11.5 IPS
 
Machine Translation (F16)
100% 616
11.3 IPS
 
Machine Translation (I8)
64% 296
5.45 IPS
 

Compare

Set Baseline Geekbench 6 now available