Geekbench AI Score

3410
Single Precision Score
3442
Half Precision Score
8257
Quantized Score
Geekbench AI 1.7.0 for Linux AVX2

Result Information

User ciassoc
Upload Date March 10 2026 11:11 AM
Views 4

System Information

AI Information
Framework OpenVINO
Backend CPU
Device Intel(R) Core(TM) Ultra 7 155H
System Information
Operating System openSUSE Tumbleweed
Model Acer Swift SFG14-72
Motherboard MTL Coral_MTH
CPU Information
Name Intel Core Ultra 7 155H
Topology 1 Processor, 16 Cores, 22 Threads
Identifier GenuineIntel Family 6 Model 170 Stepping 4
Base Frequency 4.80 GHz
Cluster 10 Cores
Memory Information
Size 31.01 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 2078
386.5 IPS
 
Image Classification (HP)
100% 2114
393.1 IPS
 
Image Classification (Q)
100% 5234
973.3 IPS
 
Image Segmentation (SP)
100% 2518
40.8 IPS
 
Image Segmentation (HP)
100% 2596
42.1 IPS
 
Image Segmentation (Q)
99% 5180
84.0 IPS
 
Pose Estimation (SP)
100% 3708
4.33 IPS
 
Pose Estimation (HP)
100% 3762
4.39 IPS
 
Pose Estimation (Q)
96% 13759
16.1 IPS
 
Object Detection (SP)
100% 1945
154.3 IPS
 
Object Detection (HP)
100% 1947
154.4 IPS
 
Object Detection (Q)
88% 5441
436.2 IPS
 
Face Detection (SP)
100% 5621
66.8 IPS
 
Face Detection (HP)
100% 5610
66.7 IPS
 
Face Detection (Q)
100% 11855
140.9 IPS
 
Depth Estimation (SP)
100% 4978
38.4 IPS
 
Depth Estimation (HP)
99% 5037
38.8 IPS
 
Depth Estimation (Q)
89% 13191
102.6 IPS
 
Style Transfer (SP)
100% 10965
14.1 IPS
 
Style Transfer (HP)
100% 11119
14.3 IPS
 
Style Transfer (Q)
98% 39263
50.6 IPS
 
Image Super-Resolution (SP)
100% 2291
84.6 IPS
 
Image Super-Resolution (HP)
100% 2312
85.4 IPS
 
Image Super-Resolution (Q)
99% 8220
304.4 IPS
 
Text Classification (SP)
100% 2218
2.96 KIPS
 
Text Classification (HP)
100% 2246
3.00 KIPS
 
Text Classification (Q)
92% 3999
5.37 KIPS
 
Machine Translation (SP)
100% 3619
62.3 IPS
 
Machine Translation (HP)
100% 3569
61.5 IPS
 
Machine Translation (Q)
100% 3599
62.0 IPS
 

Compare

Set Baseline Geekbench 6 now available