User | Phizzle |
Upload Date | August 20 2024 10:24 PM |
Views | 8 |
AI Information | |
---|---|
Framework | Core ML |
Backend | GPU |
Device | Apple A17 Pro |
System Information | |
---|---|
Operating System | iOS 17.6.1 |
Model | iPhone 15 Pro Max |
Model ID | iPhone16,2 |
Motherboard | D84AP |
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 | 7.44 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1602
297.8 IPS |
|
Image Classification (HP)
|
100% |
1399
260.2 IPS |
|
Image Classification (Q)
|
99% |
1333
248.7 IPS |
|
Image Segmentation (SP)
|
100% |
3045
49.4 IPS |
|
Image Segmentation (HP)
|
100% |
3673
59.5 IPS |
|
Image Segmentation (Q)
|
100% |
3610
58.5 IPS |
|
Pose Estimation (SP)
|
100% |
11404
13.3 IPS |
|
Pose Estimation (HP)
|
100% |
10887
12.7 IPS |
|
Pose Estimation (Q)
|
96% |
12598
14.8 IPS |
|
Object Detection (SP)
|
100% |
2429
192.7 IPS |
|
Object Detection (HP)
|
100% |
2752
218.3 IPS |
|
Object Detection (Q)
|
92% |
2358
188.2 IPS |
|
Face Detection (SP)
|
100% |
6391
75.9 IPS |
|
Face Detection (HP)
|
100% |
8466
100.6 IPS |
|
Face Detection (Q)
|
100% |
6061
72.0 IPS |
|
Depth Estimation (SP)
|
100% |
11558
89.0 IPS |
|
Depth Estimation (HP)
|
99% |
12819
98.8 IPS |
|
Depth Estimation (Q)
|
97% |
13238
102.4 IPS |
|
Style Transfer (SP)
|
100% |
29346
37.7 IPS |
|
Style Transfer (HP)
|
100% |
31097
40.0 IPS |
|
Style Transfer (Q)
|
100% |
35717
45.9 IPS |
|
Image Super-Resolution (SP)
|
100% |
6175
228.0 IPS |
|
Image Super-Resolution (HP)
|
100% |
6516
240.6 IPS |
|
Image Super-Resolution (Q)
|
98% |
5125
189.9 IPS |
|
Text Classification (SP)
|
100% |
463
618.6 IPS |
|
Text Classification (HP)
|
100% |
504
673.4 IPS |
|
Text Classification (Q)
|
93% |
310
416.2 IPS |
|
Machine Translation (SP)
|
100% |
1200
20.7 IPS |
|
Machine Translation (HP)
|
100% |
1304
22.5 IPS |
|
Machine Translation (Q)
|
98% |
1558
26.9 IPS |