Upload Date | November 13 2024 04:43 AM |
Views | 3 |
AI Information | |
---|---|
Framework | Core ML |
Backend | GPU |
Device | Apple M3 |
System Information | |
---|---|
Operating System | macOS 15.2 (Build 24C5079e) |
Model | iMac (24-inch, 2023) |
Model ID | Mac15,5 |
Motherboard | Mac15,5 |
CPU Information | |
---|---|
Name | Apple M3 |
Topology | 1 Processor, 8 Cores |
Identifier | Apple M3 |
Base Frequency | 4.05 GHz |
Cluster 1 | 4 Cores |
Cluster 2 | 4 Cores |
Memory Information | |
---|---|
Size | 16.00 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
4219
784.7 IPS |
|
Image Classification (HP)
|
100% |
4565
848.9 IPS |
|
Image Classification (Q)
|
97% |
3939
735.0 IPS |
|
Image Segmentation (SP)
|
100% |
4382
71.0 IPS |
|
Image Segmentation (HP)
|
100% |
6091
98.7 IPS |
|
Image Segmentation (Q)
|
99% |
6452
104.6 IPS |
|
Pose Estimation (SP)
|
100% |
25017
29.2 IPS |
|
Pose Estimation (HP)
|
100% |
26560
31.0 IPS |
|
Pose Estimation (Q)
|
95% |
25327
29.7 IPS |
|
Object Detection (SP)
|
100% |
4382
347.6 IPS |
|
Object Detection (HP)
|
100% |
4831
383.2 IPS |
|
Object Detection (Q)
|
91% |
4104
327.9 IPS |
|
Face Detection (SP)
|
100% |
11626
138.1 IPS |
|
Face Detection (HP)
|
100% |
14903
177.1 IPS |
|
Face Detection (Q)
|
100% |
10934
129.9 IPS |
|
Depth Estimation (SP)
|
100% |
18446
142.1 IPS |
|
Depth Estimation (HP)
|
99% |
20459
157.6 IPS |
|
Depth Estimation (Q)
|
85% |
17486
136.7 IPS |
|
Style Transfer (SP)
|
100% |
51645
66.4 IPS |
|
Style Transfer (HP)
|
100% |
60371
77.6 IPS |
|
Style Transfer (Q)
|
96% |
56559
73.0 IPS |
|
Image Super-Resolution (SP)
|
100% |
10962
404.8 IPS |
|
Image Super-Resolution (HP)
|
100% |
11658
430.5 IPS |
|
Image Super-Resolution (Q)
|
99% |
9395
346.9 IPS |
|
Text Classification (SP)
|
100% |
2844
3.80 KIPS |
|
Text Classification (HP)
|
100% |
5055
6.75 KIPS |
|
Text Classification (Q)
|
93% |
5061
6.79 KIPS |
|
Machine Translation (SP)
|
100% |
1779
30.7 IPS |
|
Machine Translation (HP)
|
100% |
1867
32.2 IPS |
|
Machine Translation (Q)
|
98% |
1658
28.6 IPS |