Geekbench ML Score

234
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.0 for Android AArch64

Result Information

Upload Date December 08 2022 02:15 PM
Views 1

System Information

System Information
Operating System Android 12
Model realme RMX3031
Model IDrealme RMX3031
Motherboard RM6893
CPU Information
Name ARM MT6893Z/CZA
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 0
Base Frequency 2.00 GHz
Cluster 14 Cores @ 2.00 GHz
Cluster 23 Cores @ 2.60 GHz
Cluster 31 Core @ 3.00 GHz
Memory Information
Size 7.33 GB
Inference Information
Framework TensorFlow Lite
Backend CPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 134
14.4 IPS
 
Image Classification (F16)
100% 114
12.3 IPS
 
Image Classification (I8)
81% 167
17.9 IPS
 
Image Segmentation (F32)
100% 429
2.65 IPS
 
Image Segmentation (F16)
100% 440
2.72 IPS
 
Image Segmentation (I8)
89% 637
3.93 IPS
 
Pose Estimation (F32)
100% 213
0.86 IPS
 
Pose Estimation (F16)
100% 202
0.82 IPS
 
Pose Estimation (I8)
80% 160
0.65 IPS
 
Object Detection (F32)
100% 122
4.28 IPS
 
Object Detection (F16)
100% 126
4.41 IPS
 
Object Detection (I8)
52% 243
8.52 IPS
 
Face Detection (F32)
100% 261
2.99 IPS
 
Face Detection (F16)
100% 256
2.94 IPS
 
Face Detection (I8)
75% 391
4.49 IPS
 
Text Classification (F32)
100% 198
78.5 IPS
 
Text Classification (F16)
100% 181
71.5 IPS
 
Machine Translation (F32)
100% 382
58.9 IPS
 
Machine Translation (F16)
100% 388
59.8 IPS
 

Compare

Set Baseline