Geekbench AI Score

109
Single Precision Score
108
Half Precision Score
78
Quantized Score
Geekbench AI 1.5.0 for Linux x86 (64-bit)

Result Information

User pingwinator
Upload Date October 17 2025 11:49 PM
Views 5

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device Intel Core i5-13600
System Information
Operating System Ubuntu 24.04.3 LTS
Model Gigabyte Technology Co., Ltd. B760M AORUS ELITE AX
Motherboard Gigabyte Technology Co., Ltd. B760M AORUS ELITE AX
CPU Information
Name Intel Core i5-13600
Topology 1 Processor, 14 Cores, 20 Threads
Identifier GenuineIntel Family 6 Model 191 Stepping 2
Base Frequency 5.00 GHz
Cluster 10 Cores
Memory Information
Size 31.11 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 66
12.2 IPS
 
Image Classification (HP)
100% 66
12.2 IPS
 
Image Classification (Q)
97% 50
9.34 IPS
 
Image Segmentation (SP)
100% 88
1.43 IPS
 
Image Segmentation (HP)
100% 88
1.42 IPS
 
Image Segmentation (Q)
98% 69
1.12 IPS
 
Pose Estimation (SP)
100% 99
0.12 IPS
 
Pose Estimation (HP)
100% 98
0.11 IPS
 
Pose Estimation (Q)
96% 84
0.10 IPS
 
Object Detection (SP)
100% 61
4.80 IPS
 
Object Detection (HP)
100% 61
4.81 IPS
 
Object Detection (Q)
87% 48
3.82 IPS
 
Face Detection (SP)
100% 202
2.40 IPS
 
Face Detection (HP)
100% 202
2.40 IPS
 
Face Detection (Q)
97% 151
1.81 IPS
 
Depth Estimation (SP)
100% 143
1.10 IPS
 
Depth Estimation (HP)
99% 140
1.08 IPS
 
Depth Estimation (Q)
64% 96
0.89 IPS
 
Style Transfer (SP)
100% 297
0.38 IPS
 
Style Transfer (HP)
100% 297
0.38 IPS
 
Style Transfer (Q)
98% 254
0.33 IPS
 
Image Super-Resolution (SP)
100% 66
2.43 IPS
 
Image Super-Resolution (HP)
100% 66
2.42 IPS
 
Image Super-Resolution (Q)
97% 52
1.91 IPS
 
Text Classification (SP)
100% 81
108.3 IPS
 
Text Classification (HP)
100% 80
107.2 IPS
 
Text Classification (Q)
91% 54
72.6 IPS
 
Machine Translation (SP)
100% 148
2.55 IPS
 
Machine Translation (HP)
100% 148
2.54 IPS
 
Machine Translation (Q)
58% 65
1.56 IPS
 

Compare

Set Baseline Geekbench 6 now available