User | justintea |
Upload Date | March 02 2025 06:40 PM |
Views | 10 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | ARM |
System Information | |
---|---|
Operating System | iOS 18.3.1 |
Model | iPhone17,5 |
Model ID | iPhone17,5 |
Motherboard | V59AP |
CPU Information | |
---|---|
Name | ARM |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 4.04 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.54 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1876
348.9 IPS |
|
Image Classification (HP)
|
100% |
15617
2.90 KIPS |
|
Image Classification (Q)
|
97% |
21045
3.93 KIPS |
|
Image Segmentation (SP)
|
100% |
1715
27.8 IPS |
|
Image Segmentation (HP)
|
100% |
18128
293.9 IPS |
|
Image Segmentation (Q)
|
99% |
23925
387.8 IPS |
|
Pose Estimation (SP)
|
100% |
7291
8.51 IPS |
|
Pose Estimation (HP)
|
100% |
136627
159.4 IPS |
|
Pose Estimation (Q)
|
95% |
278418
326.3 IPS |
|
Object Detection (SP)
|
100% |
1936
153.6 IPS |
|
Object Detection (HP)
|
100% |
16552
1.31 KIPS |
|
Object Detection (Q)
|
91% |
22719
1.82 KIPS |
|
Face Detection (SP)
|
100% |
2901
34.5 IPS |
|
Face Detection (HP)
|
100% |
41028
487.5 IPS |
|
Face Detection (Q)
|
100% |
56728
674.0 IPS |
|
Depth Estimation (SP)
|
100% |
6583
50.7 IPS |
|
Depth Estimation (HP)
|
99% |
95089
732.6 IPS |
|
Depth Estimation (Q)
|
85% |
161211
1.26 KIPS |
|
Style Transfer (SP)
|
100% |
19359
24.9 IPS |
|
Style Transfer (HP)
|
100% |
240810
309.6 IPS |
|
Style Transfer (Q)
|
96% |
305120
393.7 IPS |
|
Image Super-Resolution (SP)
|
100% |
4787
176.8 IPS |
|
Image Super-Resolution (HP)
|
100% |
49216
1.82 KIPS |
|
Image Super-Resolution (Q)
|
99% |
82846
3.06 KIPS |
|
Text Classification (SP)
|
100% |
3417
4.56 KIPS |
|
Text Classification (HP)
|
100% |
4415
5.89 KIPS |
|
Text Classification (Q)
|
93% |
4480
6.01 KIPS |
|
Machine Translation (SP)
|
100% |
3453
59.5 IPS |
|
Machine Translation (HP)
|
100% |
10270
176.9 IPS |
|
Machine Translation (Q)
|
98% |
11468
198.2 IPS |