| Upload Date | November 07 2025 04:32 AM |
| Views | 5 |
| AI Information | |
|---|---|
| Framework | Core ML |
| Backend | CPU |
| Device | Apple A17 Pro |
| System Information | |
|---|---|
| Operating System | iOS 26.1 |
| Model | iPhone 15 Pro |
| Model ID | iPhone16,1 |
| Motherboard | D83AP |
| CPU Information | |
|---|---|
| Name | Apple A17 Pro |
| Topology | 1 Processor, 6 Cores |
| Identifier | ARM |
| Base Frequency | 3.78 GHz |
| Cluster 1 | 2 Cores |
| Cluster 2 | 4 Cores |
| Memory Information | |
|---|---|
| Size | 7.50 GB |
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (SP)
|
100% |
1414
263.0 IPS |
|
|
Image Classification (HP)
|
100% |
2939
546.6 IPS |
|
|
Image Classification (Q)
|
96% |
2027
378.6 IPS |
|
|
Image Segmentation (SP)
|
100% |
1516
24.6 IPS |
|
|
Image Segmentation (HP)
|
100% |
2345
38.0 IPS |
|
|
Image Segmentation (Q)
|
99% |
2971
48.2 IPS |
|
|
Pose Estimation (SP)
|
100% |
6830
7.97 IPS |
|
|
Pose Estimation (HP)
|
100% |
15218
17.8 IPS |
|
|
Pose Estimation (Q)
|
95% |
13004
15.2 IPS |
|
|
Object Detection (SP)
|
100% |
1309
103.8 IPS |
|
|
Object Detection (HP)
|
97% |
3229
257.1 IPS |
|
|
Object Detection (Q)
|
94% |
2292
182.7 IPS |
|
|
Face Detection (SP)
|
100% |
2950
35.0 IPS |
|
|
Face Detection (HP)
|
100% |
6572
78.1 IPS |
|
|
Face Detection (Q)
|
100% |
3656
43.4 IPS |
|
|
Depth Estimation (SP)
|
100% |
6281
48.4 IPS |
|
|
Depth Estimation (HP)
|
98% |
12410
95.9 IPS |
|
|
Depth Estimation (Q)
|
85% |
8672
67.8 IPS |
|
|
Style Transfer (SP)
|
100% |
19560
25.1 IPS |
|
|
Style Transfer (HP)
|
100% |
31781
40.9 IPS |
|
|
Style Transfer (Q)
|
96% |
25905
33.4 IPS |
|
|
Image Super-Resolution (SP)
|
100% |
4342
160.3 IPS |
|
|
Image Super-Resolution (HP)
|
100% |
5734
211.7 IPS |
|
|
Image Super-Resolution (Q)
|
99% |
4285
158.2 IPS |
|
|
Text Classification (SP)
|
100% |
3199
4.27 KIPS |
|
|
Text Classification (HP)
|
99% |
5635
7.52 KIPS |
|
|
Text Classification (Q)
|
93% |
5602
7.52 KIPS |
|
|
Machine Translation (SP)
|
100% |
3125
53.8 IPS |
|
|
Machine Translation (HP)
|
98% |
7305
126.2 IPS |
|
|
Machine Translation (Q)
|
98% |
7330
126.7 IPS |