Geekbench AI Score

519
Single Precision Score
527
Half Precision Score
1091
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

Upload Date March 29 2025 04:32 PM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARM ARMv8
System Information
Operating System Android 14
Model samsung SM-M346B1
Model IDsamsung SM-M346B1
Motherboard s5e8825
Governorenergy_aware
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 1
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.40 GHz
Memory Information
Size 5.26 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 240
44.7 IPS
 
Image Classification (HP)
100% 241
44.9 IPS
 
Image Classification (Q)
100% 772
143.5 IPS
 
Image Segmentation (SP)
100% 297
4.82 IPS
 
Image Segmentation (HP)
100% 305
4.95 IPS
 
Image Segmentation (Q)
98% 864
14.0 IPS
 
Pose Estimation (SP)
100% 1145
1.34 IPS
 
Pose Estimation (HP)
100% 1153
1.35 IPS
 
Pose Estimation (Q)
84% 1783
2.11 IPS
 
Object Detection (SP)
100% 229
18.2 IPS
 
Object Detection (HP)
100% 230
18.3 IPS
 
Object Detection (Q)
83% 724
58.5 IPS
 
Face Detection (SP)
100% 727
8.64 IPS
 
Face Detection (HP)
100% 736
8.74 IPS
 
Face Detection (Q)
95% 1741
20.8 IPS
 
Depth Estimation (SP)
100% 856
6.60 IPS
 
Depth Estimation (HP)
99% 969
7.46 IPS
 
Depth Estimation (Q)
64% 1696
15.5 IPS
 
Style Transfer (SP)
89% 1929
2.50 IPS
 
Style Transfer (HP)
89% 1913
2.48 IPS
 
Style Transfer (Q)
98% 4137
5.33 IPS
 
Image Super-Resolution (SP)
100% 459
16.9 IPS
 
Image Super-Resolution (HP)
100% 455
16.8 IPS
 
Image Super-Resolution (Q)
97% 1027
38.1 IPS
 
Text Classification (SP)
100% 298
398.1 IPS
 
Text Classification (HP)
100% 297
396.3 IPS
 
Text Classification (Q)
88% 614
828.3 IPS
 
Machine Translation (SP)
100% 470
8.09 IPS
 
Machine Translation (HP)
100% 463
7.98 IPS
 
Machine Translation (Q)
50% 361
12.8 IPS
 

Compare

Set Baseline Geekbench 6 now available