Upload Date | October 20 2025 04:47 PM |
Views | 3 |
AI Information | |
---|---|
Framework | ONNX |
Backend | DirectML |
Device | NVIDIA RTX 2000 Ada Generation |
System Information | |
---|---|
Operating System | Microsoft Windows 11 Pro (64-bit) |
Model | A520M Pro4 |
Motherboard | ASRock A520M Pro4 |
CPU Information | |
---|---|
Name | AMD Ryzen 7 5700X3D |
Topology | 1 Processor, 8 Cores, 16 Threads |
Identifier | AuthenticAMD Family 25 Model 33 Stepping 2 |
Base Frequency | 3.00 GHz |
Cluster 1 | 8 Cores |
Memory Information | |
---|---|
Size | 48.00 GB |
Workload | Accuracy | Score | |
---|---|---|---|
Image Classification (SP)
|
100% |
5882
1.09 KIPS |
|
Image Classification (HP)
|
99% |
10462
1.95 KIPS |
|
Image Classification (Q)
|
99% |
5313
990.9 IPS |
|
Image Segmentation (SP)
|
100% |
9160
148.5 IPS |
|
Image Segmentation (HP)
|
100% |
12314
199.6 IPS |
|
Image Segmentation (Q)
|
98% |
8411
136.8 IPS |
|
Pose Estimation (SP)
|
100% |
64120
74.8 IPS |
|
Pose Estimation (HP)
|
100% |
203350
237.3 IPS |
|
Pose Estimation (Q)
|
96% |
57139
66.9 IPS |
|
Object Detection (SP)
|
100% |
7195
570.7 IPS |
|
Object Detection (HP)
|
100% |
11949
947.8 IPS |
|
Object Detection (Q)
|
84% |
6452
520.4 IPS |
|
Face Detection (SP)
|
100% |
16211
192.6 IPS |
|
Face Detection (HP)
|
100% |
26566
315.7 IPS |
|
Face Detection (Q)
|
97% |
14638
174.5 IPS |
|
Depth Estimation (SP)
|
100% |
31699
244.2 IPS |
|
Depth Estimation (HP)
|
99% |
62867
484.3 IPS |
|
Depth Estimation (Q)
|
77% |
26495
212.4 IPS |
|
Style Transfer (SP)
|
100% |
113903
146.4 IPS |
|
Style Transfer (HP)
|
100% |
293696
377.6 IPS |
|
Style Transfer (Q)
|
98% |
102393
132.0 IPS |
|
Image Super-Resolution (SP)
|
100% |
23832
880.0 IPS |
|
Image Super-Resolution (HP)
|
100% |
38015
1.40 KIPS |
|
Image Super-Resolution (Q)
|
99% |
17434
645.6 IPS |
|
Text Classification (SP)
|
100% |
2643
3.53 KIPS |
|
Text Classification (HP)
|
99% |
3060
4.08 KIPS |
|
Text Classification (Q)
|
97% |
1700
2.28 KIPS |
|
Machine Translation (SP)
|
100% |
2962
51.0 IPS |
|
Machine Translation (HP)
|
100% |
3185
54.9 IPS |
|
Machine Translation (Q)
|
70% |
1561
29.3 IPS |