Geekbench AI Score

283
Single Precision Score
255
Half Precision Score
536
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

User 46fy45
Upload Date October 05 2025 10:48 AM
Views 5

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device vendor
System Information
Operating System Android 12
Model HUAWEI PLA-AL10
Model IDHUAWEI PLA-AL10
Motherboard PLA
CPU Information
Name vendor
Topology 1 Processor, 8 Cores
Identifier ARM implementer 72 architecture 8 variant 2 part 3333 revision 1
Base Frequency 1.60 GHz
Cluster 14 Cores @ 1.60 GHz
Cluster 23 Cores @ 2.15 GHz
Cluster 31 Core @ 2.50 GHz
Memory Information
Size 15.28 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 236
44.0 IPS
 
Image Classification (HP)
100% 199
37.0 IPS
 
Image Classification (Q)
99% 459
85.7 IPS
 
Image Segmentation (SP)
100% 270
4.37 IPS
 
Image Segmentation (HP)
100% 148
2.39 IPS
 
Image Segmentation (Q)
98% 298
4.85 IPS
 
Pose Estimation (SP)
100% 309
0.36 IPS
 
Pose Estimation (HP)
100% 283
0.33 IPS
 
Pose Estimation (Q)
98% 710
0.83 IPS
 
Object Detection (SP)
100% 107
8.46 IPS
 
Object Detection (HP)
100% 200
15.9 IPS
 
Object Detection (Q)
87% 496
39.8 IPS
 
Face Detection (SP)
100% 520
6.17 IPS
 
Face Detection (HP)
100% 410
4.87 IPS
 
Face Detection (Q)
97% 998
11.9 IPS
 
Depth Estimation (SP)
100% 446
3.44 IPS
 
Depth Estimation (HP)
99% 447
3.44 IPS
 
Depth Estimation (Q)
64% 923
8.38 IPS
 
Style Transfer (SP)
100% 765
0.98 IPS
 
Style Transfer (HP)
100% 435
0.56 IPS
 
Style Transfer (Q)
98% 1219
1.57 IPS
 
Image Super-Resolution (SP)
100% 121
4.46 IPS
 
Image Super-Resolution (HP)
100% 121
4.46 IPS
 
Image Super-Resolution (Q)
97% 418
15.5 IPS
 
Text Classification (SP)
100% 156
208.4 IPS
 
Text Classification (HP)
100% 157
209.1 IPS
 
Text Classification (Q)
91% 272
365.3 IPS
 
Machine Translation (SP)
100% 479
8.24 IPS
 
Machine Translation (HP)
100% 477
8.22 IPS
 
Machine Translation (Q)
57% 324
7.91 IPS
 

Compare

Set Baseline Geekbench 6 now available