User | Gonfonix |
Upload Date | November 10 2024 02:10 AM |
Views | 4 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple M2 Ultra |
System Information | |
---|---|
Operating System | macOS 15.1 (Build 24B83) |
Model | Mac Pro (2023) |
Model ID | Mac14,8 |
Motherboard | Mac14,8 |
CPU Information | |
---|---|
Name | Apple M2 Ultra |
Topology | 1 Processor, 24 Cores |
Identifier | Apple M2 Ultra |
Base Frequency | 3.67 GHz |
Cluster 1 | 16 Cores |
Cluster 2 | 8 Cores |
Memory Information | |
---|---|
Size | 192.00 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
1462
271.8 IPS |
|
Image Classification (HP)
|
100% |
12935
2.41 KIPS |
|
Image Classification (Q)
|
96% |
15693
2.93 KIPS |
|
Image Segmentation (SP)
|
100% |
1781
28.9 IPS |
|
Image Segmentation (HP)
|
100% |
19408
314.6 IPS |
|
Image Segmentation (Q)
|
99% |
24345
394.7 IPS |
|
Pose Estimation (SP)
|
100% |
12306
14.4 IPS |
|
Pose Estimation (HP)
|
100% |
104963
122.5 IPS |
|
Pose Estimation (Q)
|
95% |
103496
121.3 IPS |
|
Object Detection (SP)
|
100% |
1745
138.4 IPS |
|
Object Detection (HP)
|
99% |
15704
1.25 KIPS |
|
Object Detection (Q)
|
91% |
17467
1.40 KIPS |
|
Face Detection (SP)
|
100% |
3592
42.7 IPS |
|
Face Detection (HP)
|
100% |
35492
421.7 IPS |
|
Face Detection (Q)
|
100% |
45782
544.0 IPS |
|
Depth Estimation (SP)
|
100% |
6515
50.2 IPS |
|
Depth Estimation (HP)
|
99% |
85133
655.9 IPS |
|
Depth Estimation (Q)
|
85% |
105834
827.0 IPS |
|
Style Transfer (SP)
|
100% |
27864
35.8 IPS |
|
Style Transfer (HP)
|
100% |
199496
256.5 IPS |
|
Style Transfer (Q)
|
96% |
189064
244.0 IPS |
|
Image Super-Resolution (SP)
|
100% |
6017
222.2 IPS |
|
Image Super-Resolution (HP)
|
100% |
39583
1.46 KIPS |
|
Image Super-Resolution (Q)
|
99% |
47385
1.75 KIPS |
|
Text Classification (SP)
|
100% |
2357
3.15 KIPS |
|
Text Classification (HP)
|
100% |
4071
5.43 KIPS |
|
Text Classification (Q)
|
93% |
4040
5.42 KIPS |
|
Machine Translation (SP)
|
100% |
4990
86.0 IPS |
|
Machine Translation (HP)
|
100% |
3749
64.6 IPS |
|
Machine Translation (Q)
|
98% |
3807
65.8 IPS |