User | DVB |
Upload Date | December 16 2024 03:58 PM |
Views | 4 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple A18 |
System Information | |
---|---|
Operating System | iOS 18.2 |
Model | iPhone 16 |
Model ID | iPhone17,3 |
Motherboard | D47AP |
CPU Information | |
---|---|
Name | Apple A18 |
Topology | 1 Processor, 6 Cores |
Identifier | ARM |
Base Frequency | 4.04 GHz |
Cluster 1 | 2 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 7.51 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1940
360.8 IPS |
|
Image Classification (HP)
|
100% |
15223
2.83 KIPS |
|
Image Classification (Q)
|
97% |
20370
3.80 KIPS |
|
Image Segmentation (SP)
|
100% |
1790
29.0 IPS |
|
Image Segmentation (HP)
|
100% |
18138
294.0 IPS |
|
Image Segmentation (Q)
|
99% |
23965
388.5 IPS |
|
Pose Estimation (SP)
|
100% |
8251
9.63 IPS |
|
Pose Estimation (HP)
|
100% |
149417
174.3 IPS |
|
Pose Estimation (Q)
|
95% |
294087
344.7 IPS |
|
Object Detection (SP)
|
100% |
2106
167.1 IPS |
|
Object Detection (HP)
|
100% |
16218
1.29 KIPS |
|
Object Detection (Q)
|
91% |
22258
1.78 KIPS |
|
Face Detection (SP)
|
100% |
3246
38.6 IPS |
|
Face Detection (HP)
|
100% |
40064
476.0 IPS |
|
Face Detection (Q)
|
100% |
55783
662.8 IPS |
|
Depth Estimation (SP)
|
100% |
7502
57.8 IPS |
|
Depth Estimation (HP)
|
99% |
95410
735.1 IPS |
|
Depth Estimation (Q)
|
85% |
165735
1.30 KIPS |
|
Style Transfer (SP)
|
100% |
22448
28.9 IPS |
|
Style Transfer (HP)
|
100% |
245162
315.2 IPS |
|
Style Transfer (Q)
|
96% |
302867
390.8 IPS |
|
Image Super-Resolution (SP)
|
100% |
5261
194.3 IPS |
|
Image Super-Resolution (HP)
|
100% |
49692
1.83 KIPS |
|
Image Super-Resolution (Q)
|
99% |
83544
3.08 KIPS |
|
Text Classification (SP)
|
100% |
3420
4.56 KIPS |
|
Text Classification (HP)
|
100% |
4720
6.30 KIPS |
|
Text Classification (Q)
|
93% |
4806
6.45 KIPS |
|
Machine Translation (SP)
|
100% |
3558
61.3 IPS |
|
Machine Translation (HP)
|
100% |
9437
162.5 IPS |
|
Machine Translation (Q)
|
98% |
11573
200.0 IPS |