Geekbench AI Score

169
Single Precision Score
174
Half Precision Score
120
Quantized Score
Geekbench AI 1.2.0 for Linux x86 (64-bit)

Result Information

User appstudios
Upload Date May 24 2025 04:30 PM
Views 8

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device Intel Core i9-13900HX
System Information
Operating System Arch Linux
Model Micro Computer (HK) Tech Limited Uranus Series
Motherboard Shenzhen Meigao Electronic Equipment Co.,Ltd RPFXI
CPU Information
Name Intel Core i9-13900HX
Topology 1 Processor, 24 Cores, 32 Threads
Identifier GenuineIntel Family 6 Model 183 Stepping 1
Base Frequency 5.20 GHz
Cluster 10 Cores
Memory Information
Size 94.03 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 95
17.7 IPS
 
Image Classification (HP)
100% 111
20.6 IPS
 
Image Classification (Q)
97% 79
14.7 IPS
 
Image Segmentation (SP)
100% 130
2.10 IPS
 
Image Segmentation (HP)
100% 129
2.10 IPS
 
Image Segmentation (Q)
98% 93
1.51 IPS
 
Pose Estimation (SP)
100% 149
0.17 IPS
 
Pose Estimation (HP)
100% 155
0.18 IPS
 
Pose Estimation (Q)
96% 130
0.15 IPS
 
Object Detection (SP)
100% 99
7.82 IPS
 
Object Detection (HP)
100% 101
8.02 IPS
 
Object Detection (Q)
87% 79
6.31 IPS
 
Face Detection (SP)
100% 335
3.98 IPS
 
Face Detection (HP)
100% 330
3.92 IPS
 
Face Detection (Q)
97% 248
2.96 IPS
 
Depth Estimation (SP)
100% 240
1.85 IPS
 
Depth Estimation (HP)
99% 241
1.86 IPS
 
Depth Estimation (Q)
64% 160
1.47 IPS
 
Style Transfer (SP)
100% 502
0.65 IPS
 
Style Transfer (HP)
100% 460
0.59 IPS
 
Style Transfer (Q)
98% 393
0.51 IPS
 
Image Super-Resolution (SP)
100% 104
3.84 IPS
 
Image Super-Resolution (HP)
100% 110
4.07 IPS
 
Image Super-Resolution (Q)
97% 79
2.92 IPS
 
Text Classification (SP)
100% 127
168.9 IPS
 
Text Classification (HP)
100% 127
169.3 IPS
 
Text Classification (Q)
91% 72
96.4 IPS
 
Machine Translation (SP)
100% 201
3.47 IPS
 
Machine Translation (HP)
100% 229
3.95 IPS
 
Machine Translation (Q)
58% 94
2.24 IPS
 

Compare

Set Baseline Geekbench 6 now available