Geekbench AI Score

977
Single Precision Score
1138
Half Precision Score
791
Quantized Score
Geekbench AI 1.3.0 for Windows AVX2

Result Information

Upload Date March 20 2025 02:14 PM
Views 9

System Information

AI Information
Framework ONNX
Backend DirectML
Device Intel(R) UHD Graphics 630
System Information
Operating System Microsoft Windows 10 Entreprise (64-bit)
Model Dell Inc. OptiPlex 7080
Motherboard Dell Inc. 0HV8FN
Power PlanUtilisation normale
CPU Information
Name Intel Core i5-10500
Topology 1 Processor, 6 Cores, 12 Threads
Identifier GenuineIntel Family 6 Model 165 Stepping 3
Base Frequency 3.10 GHz
Cluster 16 Cores
Memory Information
Size 16.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 401
74.6 IPS
 
Image Classification (HP)
100% 474
88.2 IPS
 
Image Classification (Q)
100% 416
77.4 IPS
 
Image Segmentation (SP)
100% 604
9.79 IPS
 
Image Segmentation (HP)
100% 708
11.5 IPS
 
Image Segmentation (Q)
98% 511
8.31 IPS
 
Pose Estimation (SP)
100% 2381
2.78 IPS
 
Pose Estimation (HP)
100% 3804
4.44 IPS
 
Pose Estimation (Q)
97% 2145
2.51 IPS
 
Object Detection (SP)
100% 503
39.9 IPS
 
Object Detection (HP)
98% 510
40.6 IPS
 
Object Detection (Q)
86% 461
37.0 IPS
 
Face Detection (SP)
100% 916
10.9 IPS
 
Face Detection (HP)
100% 1123
13.3 IPS
 
Face Detection (Q)
97% 773
9.22 IPS
 
Depth Estimation (SP)
100% 1796
13.8 IPS
 
Depth Estimation (HP)
98% 2220
17.2 IPS
 
Depth Estimation (Q)
77% 1555
12.4 IPS
 
Style Transfer (SP)
100% 6122
7.87 IPS
 
Style Transfer (HP)
100% 9127
11.7 IPS
 
Style Transfer (Q)
98% 5177
6.68 IPS
 
Image Super-Resolution (SP)
100% 1204
44.4 IPS
 
Image Super-Resolution (HP)
100% 1700
62.8 IPS
 
Image Super-Resolution (Q)
99% 936
34.7 IPS
 
Text Classification (SP)
100% 481
641.7 IPS
 
Text Classification (HP)
100% 299
399.5 IPS
 
Text Classification (Q)
97% 282
377.5 IPS
 
Machine Translation (SP)
100% 470
8.10 IPS
 
Machine Translation (HP)
100% 484
8.33 IPS
 
Machine Translation (Q)
70% 279
5.24 IPS
 

Compare

Set Baseline Geekbench 6 now available