Geekbench AI Score

180
Single Precision Score
180
Half Precision Score
417
Quantized Score
Geekbench AI 1.4.0 for Android AArch64

Result Information

User xRoach
Upload Date August 27 2025 07:05 PM
Views 5

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 15
Model samsung SM-G736U1
Model IDsamsung SM-G736U1
Motherboard lahaina
Governorschedutil
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 1
Base Frequency 1.80 GHz
Cluster 14 Cores @ 1.80 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 5.25 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 124
23.0 IPS
 
Image Classification (HP)
100% 124
23.1 IPS
 
Image Classification (Q)
99% 337
62.9 IPS
 
Image Segmentation (SP)
100% 168
2.72 IPS
 
Image Segmentation (HP)
100% 169
2.74 IPS
 
Image Segmentation (Q)
98% 375
6.09 IPS
 
Pose Estimation (SP)
100% 226
0.26 IPS
 
Pose Estimation (HP)
100% 226
0.26 IPS
 
Pose Estimation (Q)
98% 860
1.01 IPS
 
Object Detection (SP)
100% 124
9.82 IPS
 
Object Detection (HP)
100% 124
9.86 IPS
 
Object Detection (Q)
87% 363
29.1 IPS
 
Face Detection (SP)
100% 313
3.72 IPS
 
Face Detection (HP)
100% 316
3.76 IPS
 
Face Detection (Q)
97% 741
8.83 IPS
 
Depth Estimation (SP)
100% 276
2.12 IPS
 
Depth Estimation (HP)
99% 271
2.09 IPS
 
Depth Estimation (Q)
64% 504
4.58 IPS
 
Style Transfer (SP)
100% 334
0.43 IPS
 
Style Transfer (HP)
100% 334
0.43 IPS
 
Style Transfer (Q)
98% 912
1.18 IPS
 
Image Super-Resolution (SP)
100% 91
3.35 IPS
 
Image Super-Resolution (HP)
100% 91
3.36 IPS
 
Image Super-Resolution (Q)
97% 314
11.6 IPS
 
Text Classification (SP)
100% 117
156.6 IPS
 
Text Classification (HP)
100% 117
156.3 IPS
 
Text Classification (Q)
91% 218
292.7 IPS
 
Machine Translation (SP)
100% 201
3.46 IPS
 
Machine Translation (HP)
100% 202
3.48 IPS
 
Machine Translation (Q)
57% 177
4.33 IPS
 

Compare

Set Baseline Geekbench 6 now available