Geekbench ML Score

94
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.2 for Android AArch64

Result Information

User inc7007
Upload Date August 17 2024 03:14 PM
Views 4

System Information

System Information
Operating System Android 9
Model incar MAX 10
Model IDincar MAX 10
Motherboard sp9863a_2h10
Governorschedutil
CPU Information
Name ARM Unisoc
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3333 revision 0
Base Frequency 1.20 GHz
Cluster 14 Cores @ 1.20 GHz
Cluster 24 Cores @ 1.60 GHz
Memory Information
Size 2.76 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 55
5.85 IPS
 
Image Classification (F16)
100% 56
6.01 IPS
 
Image Classification (I8)
81% 209
22.4 IPS
 
Image Segmentation (F32)
100% 111
0.69 IPS
 
Image Segmentation (F16)
100% 111
0.68 IPS
 
Image Segmentation (I8)
89% 375
2.32 IPS
 
Pose Estimation (F32)
100% 69
0.28 IPS
 
Pose Estimation (F16)
100% 67
0.27 IPS
 
Pose Estimation (I8)
80% 89
0.36 IPS
 
Object Detection (F32)
100% 102
3.59 IPS
 
Object Detection (F16)
100% 105
3.67 IPS
 
Object Detection (I8)
52% 294
10.3 IPS
 
Face Detection (F32)
100% 138
1.59 IPS
 
Face Detection (F16)
100% 138
1.58 IPS
 
Face Detection (I8)
75% 318
3.65 IPS
 
Text Classification (F32)
100% 71
28.2 IPS
 
Text Classification (F16)
100% 71
28.2 IPS
 
Machine Translation (F32)
100% 17
2.64 IPS
 
Machine Translation (F16)
100% 17
2.64 IPS
 

Compare

Set Baseline Geekbench 6 now available