User | mrgnw |
Upload Date | December 26 2024 07:16 AM |
Views | 4 |
AI Information | |
---|---|
Framework | Core ML |
Backend | Neural Engine |
Device | Apple M2 Ultra |
System Information | |
---|---|
Operating System | macOS 15.2 (Build 24C101) |
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% |
1467
272.9 IPS |
|
Image Classification (HP)
|
100% |
13807
2.57 KIPS |
|
Image Classification (Q)
|
96% |
16203
3.03 KIPS |
|
Image Segmentation (SP)
|
100% |
1801
29.2 IPS |
|
Image Segmentation (HP)
|
100% |
19749
320.2 IPS |
|
Image Segmentation (Q)
|
99% |
24604
398.9 IPS |
|
Pose Estimation (SP)
|
100% |
18952
22.1 IPS |
|
Pose Estimation (HP)
|
100% |
105497
123.1 IPS |
|
Pose Estimation (Q)
|
95% |
104307
122.3 IPS |
|
Object Detection (SP)
|
100% |
1842
146.1 IPS |
|
Object Detection (HP)
|
99% |
15871
1.26 KIPS |
|
Object Detection (Q)
|
91% |
17538
1.40 KIPS |
|
Face Detection (SP)
|
100% |
3936
46.8 IPS |
|
Face Detection (HP)
|
100% |
34536
410.4 IPS |
|
Face Detection (Q)
|
100% |
44362
527.1 IPS |
|
Depth Estimation (SP)
|
100% |
6502
50.1 IPS |
|
Depth Estimation (HP)
|
99% |
85572
659.3 IPS |
|
Depth Estimation (Q)
|
85% |
109843
858.4 IPS |
|
Style Transfer (SP)
|
100% |
27084
34.8 IPS |
|
Style Transfer (HP)
|
100% |
200968
258.3 IPS |
|
Style Transfer (Q)
|
96% |
189991
245.2 IPS |
|
Image Super-Resolution (SP)
|
100% |
5884
217.3 IPS |
|
Image Super-Resolution (HP)
|
100% |
40613
1.50 KIPS |
|
Image Super-Resolution (Q)
|
99% |
48452
1.79 KIPS |
|
Text Classification (SP)
|
100% |
2338
3.12 KIPS |
|
Text Classification (HP)
|
100% |
3978
5.31 KIPS |
|
Text Classification (Q)
|
93% |
3965
5.32 KIPS |
|
Machine Translation (SP)
|
100% |
4895
84.3 IPS |
|
Machine Translation (HP)
|
100% |
4016
69.2 IPS |
|
Machine Translation (Q)
|
98% |
4198
72.5 IPS |