Geekbench AI Score

542
Single Precision Score
1322
Half Precision Score
2713
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

Upload Date January 07 2025 05:23 AM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 14
Model motorola motorola edge 50 neo
Model IDmotorola motorola edge 50 neo
Motherboard vienna
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 0
Base Frequency 2.00 GHz
Cluster 14 Cores @ 2.00 GHz
Cluster 24 Cores @ 2.50 GHz
Memory Information
Size 11.17 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 436
81.1 IPS
 
Image Classification (HP)
100% 1520
282.7 IPS
 
Image Classification (Q)
99% 4581
854.5 IPS
 
Image Segmentation (SP)
100% 559
9.06 IPS
 
Image Segmentation (HP)
100% 1625
26.3 IPS
 
Image Segmentation (Q)
98% 2032
33.0 IPS
 
Pose Estimation (SP)
100% 2677
3.12 IPS
 
Pose Estimation (HP)
100% 6676
7.79 IPS
 
Pose Estimation (Q)
95% 27018
31.7 IPS
 
Object Detection (SP)
100% 311
24.7 IPS
 
Object Detection (HP)
100% 1402
111.2 IPS
 
Object Detection (Q)
86% 2225
179.0 IPS
 
Face Detection (SP)
100% 1301
15.5 IPS
 
Face Detection (HP)
100% 3769
44.8 IPS
 
Face Detection (Q)
97% 9220
110.0 IPS
 
Depth Estimation (SP)
100% 400
3.09 IPS
 
Depth Estimation (HP)
99% 6996
53.9 IPS
 
Depth Estimation (Q)
62% 14567
138.3 IPS
 
Style Transfer (SP)
100% 1683
2.16 IPS
 
Style Transfer (HP)
100% 305
0.39 IPS
 
Style Transfer (Q)
30% 381
5.75 IPS
 
Image Super-Resolution (SP)
100% 930
34.3 IPS
 
Image Super-Resolution (HP)
100% 4168
153.9 IPS
 
Image Super-Resolution (Q)
41% 3081
448.7 IPS
 
Text Classification (SP)
100% 72
96.6 IPS
 
Text Classification (HP)
100% 112
149.5 IPS
 
Text Classification (Q)
93% 1321
1.77 KIPS
 
Machine Translation (SP)
100% 184
3.18 IPS
 
Machine Translation (HP)
100% 188
3.24 IPS
 
Machine Translation (Q)
53% 186
5.36 IPS
 

Compare

Set Baseline Geekbench 6 now available