User | non |
Upload Date | September 20 2024 10:52 AM |
Views | 8 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple M2 Ultra |
System Information | |
---|---|
Operating System | macOS 15.1 (Build 24B5046f) |
Model | Mac Studio (2023) |
Model ID | Mac14,14 |
Motherboard | Mac14,14 |
CPU Information | |
---|---|
Name | Apple M2 Ultra |
Topology | 1 Processor, 24 Cores |
Identifier | Apple M2 Ultra |
Base Frequency | 3.48 GHz |
Cluster 1 | 16 Cores |
Cluster 2 | 8 Cores |
Memory Information | |
---|---|
Size | 64.00 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1450
269.7 IPS |
|
Image Classification (HP)
|
100% |
12578
2.34 KIPS |
|
Image Classification (Q)
|
99% |
15112
2.82 KIPS |
|
Image Segmentation (SP)
|
100% |
1774
28.8 IPS |
|
Image Segmentation (HP)
|
100% |
9286
150.5 IPS |
|
Image Segmentation (Q)
|
100% |
9281
150.5 IPS |
|
Pose Estimation (SP)
|
100% |
18118
21.1 IPS |
|
Pose Estimation (HP)
|
100% |
102431
119.5 IPS |
|
Pose Estimation (Q)
|
94% |
96535
113.2 IPS |
|
Object Detection (SP)
|
100% |
1842
146.1 IPS |
|
Object Detection (HP)
|
99% |
14146
1.13 KIPS |
|
Object Detection (Q)
|
91% |
16945
1.35 KIPS |
|
Face Detection (SP)
|
100% |
3676
43.7 IPS |
|
Face Detection (HP)
|
100% |
23679
281.4 IPS |
|
Face Detection (Q)
|
100% |
36600
434.9 IPS |
|
Depth Estimation (SP)
|
100% |
6485
50.0 IPS |
|
Depth Estimation (HP)
|
99% |
72008
554.8 IPS |
|
Depth Estimation (Q)
|
97% |
81110
627.2 IPS |
|
Style Transfer (SP)
|
100% |
27366
35.2 IPS |
|
Style Transfer (HP)
|
100% |
155439
199.8 IPS |
|
Style Transfer (Q)
|
100% |
157638
202.6 IPS |
|
Image Super-Resolution (SP)
|
100% |
5956
219.9 IPS |
|
Image Super-Resolution (HP)
|
100% |
38921
1.44 KIPS |
|
Image Super-Resolution (Q)
|
98% |
46935
1.74 KIPS |
|
Text Classification (SP)
|
100% |
2222
2.97 KIPS |
|
Text Classification (HP)
|
100% |
3838
5.12 KIPS |
|
Text Classification (Q)
|
93% |
3837
5.15 KIPS |
|
Machine Translation (SP)
|
100% |
4920
84.7 IPS |
|
Machine Translation (HP)
|
100% |
2023
34.8 IPS |
|
Machine Translation (Q)
|
98% |
2014
34.8 IPS |