Geekbench ML Score

489
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.0 for Android AArch64

Result Information

Upload Date July 16 2022 08:23 PM
Views 1

System Information

System Information
Operating System Android 12
Model samsung SM-S908U1
Model IDsamsung SM-S908U1
Motherboard taro
Governorwalt
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 2 part 3400 revision 0
Base Frequency 1.78 GHz
Cluster 14 Cores @ 1.79 GHz
Cluster 23 Cores @ 2.50 GHz
Cluster 31 Core @ 3.00 GHz
Memory Information
Size 7.05 GB
Inference Information
Framework TensorFlow Lite
Backend CPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 476
51.1 IPS
 
Image Classification (F16)
100% 466
49.9 IPS
 
Image Classification (I8)
81% 193
20.7 IPS
 
Image Segmentation (F32)
100% 767
4.74 IPS
 
Image Segmentation (F16)
100% 755
4.66 IPS
 
Image Segmentation (I8)
89% 887
5.48 IPS
 
Pose Estimation (F32)
100% 287
1.16 IPS
 
Pose Estimation (F16)
100% 280
1.13 IPS
 
Pose Estimation (I8)
80% 232
0.94 IPS
 
Object Detection (F32)
100% 492
17.3 IPS
 
Object Detection (F16)
100% 510
17.9 IPS
 
Object Detection (I8)
52% 307
10.8 IPS
 
Face Detection (F32)
100% 767
8.81 IPS
 
Face Detection (F16)
100% 762
8.75 IPS
 
Face Detection (I8)
75% 540
6.21 IPS
 
Text Classification (F32)
100% 343
135.5 IPS
 
Text Classification (F16)
100% 335
132.4 IPS
 
Machine Translation (F32)
100% 1063
163.9 IPS
 
Machine Translation (F16)
100% 1064
164.0 IPS
 

Compare

Set Baseline