Geekbench ML Score

1639
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date November 26 2024 05:07 PM
Views 1

System Information

System Information
Operating System Android 15
Model OnePlus PJZ110
Model IDOnePlus PJZ110
Motherboard sun
Governorwalt
CPU Information
Name Qualcomm ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 3 part 1 revision 4
Base Frequency 3.53 GHz
Cluster 16 Cores @ 3.53 GHz
Cluster 22 Cores @ 4.32 GHz
Memory Information
Size 14.75 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 926
173.3 IPS
 
Image Classification (F16)
100% 919
172.0 IPS
 
Image Classification (I8)
97% 1553
290.6 IPS
 
Image Segmentation (F32)
100% 1466
24.5 IPS
 
Image Segmentation (F16)
100% 1396
23.3 IPS
 
Image Segmentation (I8)
98% 1387
23.2 IPS
 
Pose Estimation (F32)
100% 1825
2.21 IPS
 
Pose Estimation (F16)
100% 1850
2.24 IPS
 
Pose Estimation (I8)
100% 8619
10.4 IPS
 
Object Detection (F32)
100% 1113
83.1 IPS
 
Object Detection (F16)
100% 926
69.1 IPS
 
Object Detection (I8)
61% 2088
155.9 IPS
 
Face Detection (F32)
100% 1958
23.3 IPS
 
Face Detection (F16)
100% 1975
23.5 IPS
 
Face Detection (I8)
86% 3905
46.4 IPS
 
Depth Estimation (F32)
100% 1978
15.3 IPS
 
Depth Estimation (F16)
100% 1976
15.3 IPS
 
Depth Estimation (I8)
95% 3905
30.3 IPS
 
Style Transfer (F32)
100% 3112
4.09 IPS
 
Style Transfer (F16)
100% 3087
4.06 IPS
 
Style Transfer (I8)
98% 8037
10.6 IPS
 
Image Super-Resolution (F32)
100% 1193
42.6 IPS
 
Image Super-Resolution (F16)
100% 1256
44.9 IPS
 
Image Super-Resolution (I8)
98% 3765
134.5 IPS
 
Text Classification (F32)
100% 492
706.5 IPS
 
Text Classification (F16)
100% 513
737.5 IPS
 
Text Classification (I8)
92% 784
1.13 KIPS
 
Machine Translation (F32)
100% 834
15.3 IPS
 
Machine Translation (F16)
100% 897
16.5 IPS
 
Machine Translation (I8)
64% 793
14.6 IPS
 

Compare

Set Baseline Geekbench 6 now available