Geekbench AI Score

1314
Single Precision Score
704
Half Precision Score
1995
Quantized Score
Geekbench AI 1.2.0 for Windows AVX2

Result Information

User sso7746
Upload Date February 24 2025 03:50 PM
Views 10

System Information

AI Information
Framework ONNX
Backend CPU
Device Intel Core i3-10105
System Information
Operating System Microsoft Windows 11 Enterprise LTSC (64-bit)
Model innotek GmbH VirtualBox
Motherboard Oracle Corporation VirtualBox
Power PlanBalanced
CPU Information
Name Intel Core i3-10105
Topology 1 Processor, 4 Cores
Identifier GenuineIntel Family 6 Model 165 Stepping 3
Base Frequency 3.70 GHz
Cluster 14 Cores
Memory Information
Size

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 996
185.3 IPS
 
Image Classification (HP)
100% 293
54.6 IPS
 
Image Classification (Q)
96% 1767
330.0 IPS
 
Image Segmentation (SP)
100% 496
8.04 IPS
 
Image Segmentation (HP)
100% 396
6.42 IPS
 
Image Segmentation (Q)
97% 1279
20.8 IPS
 
Pose Estimation (SP)
100% 2242
2.62 IPS
 
Pose Estimation (HP)
100% 1885
2.20 IPS
 
Pose Estimation (Q)
86% 4566
5.40 IPS
 
Object Detection (SP)
100% 922
73.1 IPS
 
Object Detection (HP)
100% 339
26.9 IPS
 
Object Detection (Q)
58% 1262
137.9 IPS
 
Face Detection (SP)
100% 1375
16.3 IPS
 
Face Detection (HP)
100% 649
7.71 IPS
 
Face Detection (Q)
96% 2983
35.6 IPS
 
Depth Estimation (SP)
100% 2731
21.0 IPS
 
Depth Estimation (HP)
99% 1080
8.32 IPS
 
Depth Estimation (Q)
78% 4077
32.5 IPS
 
Style Transfer (SP)
100% 5331
6.85 IPS
 
Style Transfer (HP)
100% 4531
5.82 IPS
 
Style Transfer (Q)
87% 5140
6.69 IPS
 
Image Super-Resolution (SP)
100% 1109
41.0 IPS
 
Image Super-Resolution (HP)
100% 815
30.1 IPS
 
Image Super-Resolution (Q)
97% 1543
57.2 IPS
 
Text Classification (SP)
100% 803
1.07 KIPS
 
Text Classification (HP)
100% 460
614.1 IPS
 
Text Classification (Q)
97% 843
1.13 KIPS
 
Machine Translation (SP)
100% 849
14.6 IPS
 
Machine Translation (HP)
100% 343
5.91 IPS
 
Machine Translation (Q)
68% 944
18.2 IPS
 

Compare

Set Baseline Geekbench 6 now available