Geekbench AI Score

790
Single Precision Score
1136
Half Precision Score
1068
Quantized Score
Geekbench AI 1.5.0 for Windows x86 (64-bit)

Result Information

Upload Date November 20 2025 12:35 PM
Views 2

System Information

AI Information
Framework OpenVINO
Backend GPU
Device Intel(R) HD Graphics 615 (iGPU)
System Information
Operating System Microsoft Windows 10 (64-bit)
Model Microsoft Corporation Surface Go
Motherboard Microsoft Corporation Surface Go
CPU Information
Name Intel Pentium 4415Y
Topology 1 Processor, 2 Cores, 4 Threads
Identifier GenuineIntel Family 6 Model 142 Stepping 9
Base Frequency 1.61 GHz
Cluster 12 Cores
Memory Information
Size 8.00 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 521
96.9 IPS
 
Image Classification (HP)
100% 568
105.6 IPS
 
Image Classification (Q)
100% 516
95.9 IPS
 
Image Segmentation (SP)
100% 700
11.3 IPS
 
Image Segmentation (HP)
100% 1188
19.3 IPS
 
Image Segmentation (Q)
99% 1101
17.9 IPS
 
Pose Estimation (SP)
100% 1388
1.62 IPS
 
Pose Estimation (HP)
99% 1475
1.73 IPS
 
Pose Estimation (Q)
97% 1968
2.30 IPS
 
Object Detection (SP)
100% 504
40.0 IPS
 
Object Detection (HP)
99% 732
58.3 IPS
 
Object Detection (Q)
88% 633
50.8 IPS
 
Face Detection (SP)
100% 1489
17.7 IPS
 
Face Detection (HP)
100% 2392
28.4 IPS
 
Face Detection (Q)
97% 2257
26.9 IPS
 
Depth Estimation (SP)
100% 1645
12.7 IPS
 
Depth Estimation (HP)
98% 2883
22.3 IPS
 
Depth Estimation (Q)
89% 2607
20.3 IPS
 
Style Transfer (SP)
100% 3900
5.01 IPS
 
Style Transfer (HP)
100% 5198
6.68 IPS
 
Style Transfer (Q)
98% 4701
6.06 IPS
 
Image Super-Resolution (SP)
100% 684
25.2 IPS
 
Image Super-Resolution (HP)
100% 1217
44.9 IPS
 
Image Super-Resolution (Q)
99% 1147
42.5 IPS
 
Text Classification (SP)
71% 216
312.8 IPS
 
Text Classification (HP)
71% 232
335.9 IPS
 
Text Classification (Q)
92% 179
240.2 IPS
 
Machine Translation (SP)
100% 266
4.59 IPS
 
Machine Translation (HP)
99% 486
8.39 IPS
 
Machine Translation (Q)
98% 481
8.30 IPS
 

Compare

Set Baseline Geekbench 6 now available