Geekbench ML Score

102
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.0 for Android AArch64

Result Information

Upload Date December 08 2022 07:43 PM
Views 1

System Information

System Information
Operating System Android 11
Model Google octopus
Model IDGoogle octopus
Motherboard bobba
CPU Information
Name placeholder
Topology 1 Processor, 4 Cores
Identifier ARM implementer 78 architecture 8 variant 2 part 0 revision 1
Base Frequency 0 MHz
Cluster 14 Cores @ 0.00 Hz
Memory Information
Size 2.70 GB
Inference Information
Framework TensorFlow Lite
Backend CPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 102
10.9 IPS
 
Image Classification (F16)
100% 103
11.0 IPS
 
Image Classification (I8)
81% 76
8.16 IPS
 
Image Segmentation (F32)
100% 120
0.74 IPS
 
Image Segmentation (F16)
100% 121
0.75 IPS
 
Image Segmentation (I8)
89% 151
0.94 IPS
 
Pose Estimation (F32)
100% 52
0.21 IPS
 
Pose Estimation (F16)
100% 50
0.20 IPS
 
Pose Estimation (I8)
80% 37
0.15 IPS
 
Object Detection (F32)
100% 128
4.50 IPS
 
Object Detection (F16)
100% 127
4.45 IPS
 
Object Detection (I8)
52% 101
3.53 IPS
 
Face Detection (F32)
100% 165
1.90 IPS
 
Face Detection (F16)
100% 166
1.90 IPS
 
Face Detection (I8)
75% 101
1.15 IPS
 
Text Classification (F32)
100% 103
40.6 IPS
 
Text Classification (F16)
100% 104
41.2 IPS
 
Machine Translation (F32)
100% 142
21.9 IPS
 
Machine Translation (F16)
100% 145
22.4 IPS
 

Compare

Set Baseline