Geekbench AI Score

1056
Single Precision Score
1048
Half Precision Score
1767
Quantized Score
Geekbench AI 1.5.0 for Android AArch64

Result Information

Upload Date December 07 2025 08:00 AM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARMv8
System Information
Operating System Android 12
Model HUAWEI HBP-LX9
Model IDHUAWEI HBP-LX9
Motherboard HBP
CPU Information
Name ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 72 architecture 8 variant 2 part 3331 revision 0
Base Frequency 1.55 GHz
Cluster 14 Cores @ 1.55 GHz
Cluster 23 Cores @ 2.18 GHz
Cluster 31 Core @ 2.30 GHz
Memory Information
Size 15.23 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 493
91.6 IPS
 
Image Classification (HP)
100% 485
90.2 IPS
 
Image Classification (Q)
100% 1381
256.7 IPS
 
Image Segmentation (SP)
100% 638
10.3 IPS
 
Image Segmentation (HP)
100% 631
10.2 IPS
 
Image Segmentation (Q)
98% 1497
24.3 IPS
 
Pose Estimation (SP)
100% 2271
2.65 IPS
 
Pose Estimation (HP)
100% 2360
2.75 IPS
 
Pose Estimation (Q)
84% 2839
3.37 IPS
 
Object Detection (SP)
98% 467
37.1 IPS
 
Object Detection (HP)
98% 439
34.9 IPS
 
Object Detection (Q)
83% 1251
101.1 IPS
 
Face Detection (SP)
100% 1445
17.2 IPS
 
Face Detection (HP)
100% 1483
17.6 IPS
 
Face Detection (Q)
95% 3158
37.7 IPS
 
Depth Estimation (SP)
99% 2126
16.4 IPS
 
Depth Estimation (HP)
99% 1982
15.3 IPS
 
Depth Estimation (Q)
64% 2798
25.6 IPS
 
Style Transfer (SP)
89% 3572
4.64 IPS
 
Style Transfer (HP)
89% 3526
4.58 IPS
 
Style Transfer (Q)
98% 6237
8.04 IPS
 
Image Super-Resolution (SP)
100% 1192
44.0 IPS
 
Image Super-Resolution (HP)
100% 1209
44.6 IPS
 
Image Super-Resolution (Q)
97% 1699
62.9 IPS
 
Text Classification (SP)
100% 543
724.8 IPS
 
Text Classification (HP)
99% 552
737.1 IPS
 
Text Classification (Q)
88% 930
1.25 KIPS
 
Machine Translation (SP)
100% 734
12.7 IPS
 
Machine Translation (HP)
100% 732
12.6 IPS
 
Machine Translation (Q)
50% 465
16.5 IPS
 

Compare

Set Baseline Geekbench 6 now available