User | Tom34772 |
Upload Date | September 20 2025 11:45 PM |
Views | 4 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 26.0 |
Model | iPhone18,2 |
Model ID | iPhone18,2 |
Motherboard | V54AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 4.25 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 11.34 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
2341
435.4 IPS |
|
Image Classification (HP)
|
100% |
16076
2.99 KIPS |
|
Image Classification (Q)
|
97% |
22246
4.15 KIPS |
|
Image Segmentation (SP)
|
100% |
2131
34.5 IPS |
|
Image Segmentation (HP)
|
100% |
20005
324.3 IPS |
|
Image Segmentation (Q)
|
99% |
25985
421.2 IPS |
|
Pose Estimation (SP)
|
100% |
10646
12.4 IPS |
|
Pose Estimation (HP)
|
100% |
155023
180.9 IPS |
|
Pose Estimation (Q)
|
95% |
290016
339.9 IPS |
|
Object Detection (SP)
|
100% |
2687
213.2 IPS |
|
Object Detection (HP)
|
100% |
17335
1.38 KIPS |
|
Object Detection (Q)
|
91% |
25118
2.01 KIPS |
|
Face Detection (SP)
|
100% |
4591
54.6 IPS |
|
Face Detection (HP)
|
100% |
40299
478.8 IPS |
|
Face Detection (Q)
|
100% |
51672
614.0 IPS |
|
Depth Estimation (SP)
|
100% |
8487
65.4 IPS |
|
Depth Estimation (HP)
|
99% |
104853
807.8 IPS |
|
Depth Estimation (Q)
|
85% |
167994
1.31 KIPS |
|
Style Transfer (SP)
|
100% |
24902
32.0 IPS |
|
Style Transfer (HP)
|
100% |
259049
333.0 IPS |
|
Style Transfer (Q)
|
96% |
351815
454.0 IPS |
|
Image Super-Resolution (SP)
|
100% |
5647
208.5 IPS |
|
Image Super-Resolution (HP)
|
100% |
56160
2.07 KIPS |
|
Image Super-Resolution (Q)
|
99% |
85451
3.16 KIPS |
|
Text Classification (SP)
|
100% |
3240
4.32 KIPS |
|
Text Classification (HP)
|
99% |
6097
8.14 KIPS |
|
Text Classification (Q)
|
93% |
6084
8.17 KIPS |
|
Machine Translation (SP)
|
100% |
5147
88.7 IPS |
|
Machine Translation (HP)
|
100% |
11239
193.6 IPS |
|
Machine Translation (Q)
|
98% |
12489
215.8 IPS |