Geekbench ML Score

363
TensorFlow Lite NNAPI Inference Score
Geekbench ML 0.5.0 for Android AArch64

Result Information

Upload Date May 14 2022 11:25 AM
Views 1

System Information

System Information
Operating System Android 12
Model Google Pixel 4a
Model IDGoogle Pixel 4a
Motherboard sunfish
Governorschedutil
CPU Information
Name Qualcomm Qualcomm
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 13 part 2052 revision 14
Base Frequency 1.80 GHz
Cluster 16 Cores @ 1.80 GHz
Cluster 22 Cores @ 2.21 GHz
Memory Information
Size 5.46 GB
Inference Information
Framework TensorFlow Lite
Backend NNAPI

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 309
33.1 IPS
 
Image Classification (F16)
100% 429
46.0 IPS
 
Image Classification (I8)
80% 1190
127.5 IPS
 
Image Segmentation (F32)
100% 446
2.76 IPS
 
Image Segmentation (F16)
100% 544
3.36 IPS
 
Image Segmentation (I8)
89% 2365
14.6 IPS
 
Pose Estimation (F32)
100% 136
0.55 IPS
 
Pose Estimation (F16)
100% 318
1.29 IPS
 
Pose Estimation (I8)
80% 1267
5.13 IPS
 
Object Detection (F32)
100% 306
10.8 IPS
 
Object Detection (F16)
100% 417
14.6 IPS
 
Object Detection (I8)
52% 661
23.2 IPS
 
Face Detection (F32)
100% 416
4.78 IPS
 
Face Detection (F16)
97% 504
5.79 IPS
 
Face Detection (I8)
75% 1828
21.0 IPS
 
Text Classification (F32)
54% 24
9.68 IPS
 
Text Classification (F16)
54% 24
9.38 IPS
 
Machine Translation (F32)
100% 219
33.8 IPS
 
Machine Translation (F16)
100% 219
33.8 IPS
 

Compare

Set Baseline

Share

Geekbench ML now available for download