| Upload Date | January 05 2026 06:51 PM |
| Views | 6 |
| AI Information | |
|---|---|
| Framework | TensorFlow Lite |
| Backend | GPU |
| Device | ARMv8 |
| System Information | |
|---|---|
| Operating System | Android 12 |
| Model | HUAWEI HBP-LX9 |
| Model ID | HUAWEI 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 1 | 4 Cores @ 1.55 GHz |
| Cluster 2 | 3 Cores @ 2.18 GHz |
| Cluster 3 | 1 Core @ 2.30 GHz |
| Memory Information | |
|---|---|
| Size | 15.23 GB |
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (SP)
|
100% |
107
19.8 IPS |
|
|
Image Classification (HP)
|
100% |
128
23.8 IPS |
|
|
Image Classification (Q)
|
100% |
130
24.2 IPS |
|
|
Image Segmentation (SP)
|
100% |
151
2.44 IPS |
|
|
Image Segmentation (HP)
|
100% |
201
3.26 IPS |
|
|
Image Segmentation (Q)
|
98% |
197
3.20 IPS |
|
|
Pose Estimation (SP)
|
100% |
724
0.84 IPS |
|
|
Pose Estimation (HP)
|
99% |
1049
1.22 IPS |
|
|
Pose Estimation (Q)
|
95% |
1037
1.21 IPS |
|
|
Object Detection (SP)
|
100% |
90
7.13 IPS |
|
|
Object Detection (HP)
|
99% |
107
8.52 IPS |
|
|
Object Detection (Q)
|
85% |
100
8.05 IPS |
|
|
Face Detection (SP)
|
100% |
373
4.44 IPS |
|
|
Face Detection (HP)
|
100% |
517
6.14 IPS |
|
|
Face Detection (Q)
|
97% |
497
5.93 IPS |
|
|
Depth Estimation (SP)
|
100% |
422
3.25 IPS |
|
|
Depth Estimation (HP)
|
98% |
583
4.51 IPS |
|
|
Depth Estimation (Q)
|
61% |
467
4.62 IPS |
|
|
Style Transfer (SP)
|
100% |
1074
1.38 IPS |
|
|
Style Transfer (HP)
|
100% |
1578
2.03 IPS |
|
|
Style Transfer (Q)
|
98% |
1565
2.02 IPS |
|
|
Image Super-Resolution (SP)
|
100% |
258
9.53 IPS |
|
|
Image Super-Resolution (HP)
|
100% |
357
13.2 IPS |
|
|
Image Super-Resolution (Q)
|
97% |
393
14.6 IPS |
|
|
Text Classification (SP)
|
35% |
7
64.7 IPS |
|
|
Text Classification (HP)
|
35% |
8
71.6 IPS |
|
|
Text Classification (Q)
|
33% |
6
71.1 IPS |
|
|
Machine Translation (SP)
|
100% |
79
1.37 IPS |
|
|
Machine Translation (HP)
|
100% |
86
1.48 IPS |
|
|
Machine Translation (Q)
|
48% |
36
1.37 IPS |