Upload Date | December 28 2024 01:23 PM |
Views | 1 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple A16 Bionic |
System Information | |
---|---|
Operating System | iOS 18.2 |
Model | iPhone 15 |
Model ID | iPhone15,4 |
Motherboard | D37AP |
CPU Information | |
---|---|
Name | Apple A16 Bionic |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 3.46 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 5.56 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1379
256.5 IPS |
|
Image Classification (HP)
|
100% |
9830
1.83 KIPS |
|
Image Classification (Q)
|
97% |
12614
2.35 KIPS |
|
Image Segmentation (SP)
|
100% |
1481
24.0 IPS |
|
Image Segmentation (HP)
|
100% |
14283
231.5 IPS |
|
Image Segmentation (Q)
|
99% |
16960
274.9 IPS |
|
Pose Estimation (SP)
|
100% |
6756
7.88 IPS |
|
Pose Estimation (HP)
|
100% |
98742
115.2 IPS |
|
Pose Estimation (Q)
|
95% |
102064
119.6 IPS |
|
Object Detection (SP)
|
100% |
1603
127.1 IPS |
|
Object Detection (HP)
|
100% |
12756
1.01 KIPS |
|
Object Detection (Q)
|
91% |
14808
1.18 KIPS |
|
Face Detection (SP)
|
100% |
3211
38.2 IPS |
|
Face Detection (HP)
|
100% |
31327
372.2 IPS |
|
Face Detection (Q)
|
100% |
39643
471.0 IPS |
|
Depth Estimation (SP)
|
100% |
6090
46.9 IPS |
|
Depth Estimation (HP)
|
99% |
74859
576.7 IPS |
|
Depth Estimation (Q)
|
85% |
95837
749.3 IPS |
|
Style Transfer (SP)
|
100% |
18355
23.6 IPS |
|
Style Transfer (HP)
|
100% |
171208
220.1 IPS |
|
Style Transfer (Q)
|
96% |
176534
227.8 IPS |
|
Image Super-Resolution (SP)
|
100% |
4113
151.9 IPS |
|
Image Super-Resolution (HP)
|
100% |
31745
1.17 KIPS |
|
Image Super-Resolution (Q)
|
99% |
36981
1.37 KIPS |
|
Text Classification (SP)
|
100% |
2615
3.49 KIPS |
|
Text Classification (HP)
|
100% |
2178
2.91 KIPS |
|
Text Classification (Q)
|
93% |
1979
2.66 KIPS |
|
Machine Translation (SP)
|
100% |
3394
58.5 IPS |
|
Machine Translation (HP)
|
100% |
7534
129.8 IPS |
|
Machine Translation (Q)
|
98% |
8096
139.9 IPS |