Geekbench ML Score

926
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.0 for Android AArch64

Result Information

Upload Date July 16 2022 05:32 PM
Views 1

System Information

System Information
Operating System Android 12
Model realme RMX3363
Model IDrealme RMX3363
Motherboard lahaina
Governorschedutil
CPU Information
Name ARM Qualcomm
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 5.16 GB
Inference Information
Framework TensorFlow Lite
Backend GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 701
75.2 IPS
 
Image Classification (F16)
100% 1012
108.4 IPS
 
Image Classification (I8)
80% 886
94.9 IPS
 
Image Segmentation (F32)
100% 1526
9.43 IPS
 
Image Segmentation (F16)
100% 2587
16.0 IPS
 
Image Segmentation (I8)
89% 2615
16.1 IPS
 
Pose Estimation (F32)
100% 688
2.78 IPS
 
Pose Estimation (F16)
100% 1336
5.40 IPS
 
Pose Estimation (I8)
80% 1340
5.42 IPS
 
Object Detection (F32)
100% 469
16.5 IPS
 
Object Detection (F16)
99% 604
21.2 IPS
 
Object Detection (I8)
53% 447
15.7 IPS
 
Face Detection (F32)
100% 1185
13.6 IPS
 
Face Detection (F16)
97% 2048
23.5 IPS
 
Face Detection (I8)
75% 1665
19.1 IPS
 
Text Classification (F32)
100% 426
168.4 IPS
 
Text Classification (F16)
98% 569
224.9 IPS
 
Machine Translation (F32)
100% 484
74.6 IPS
 
Machine Translation (F16)
100% 486
74.9 IPS
 

Compare

Set Baseline