Geekbench AI Score

372
Single Precision Score
364
Half Precision Score
159
Quantized Score
Geekbench AI 1.0.0 for Android AArch64

Result Information

Upload Date August 24 2024 05:56 AM
Views 15

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 14
Model Amlogic anemone
Model IDAmlogic anemone
Motherboard anemone
Governorschedutil
CPU Information
Name ARM ARMv8
Topology 1 Processor, 5 Cores
Identifier ARM implementer 65 architecture 8 variant 4 part 3339 revision 1
Base Frequency 1.86 GHz
Cluster 15 Cores @ 1.86 GHz
Memory Information
Size 7.68 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 307
57.1 IPS
 
Image Classification (HP)
100% 250
46.5 IPS
 
Image Classification (Q)
97% 1308
244.1 IPS
 
Image Segmentation (SP)
100% 388
6.29 IPS
 
Image Segmentation (HP)
100% 386
6.26 IPS
 
Image Segmentation (Q)
98% 756
12.3 IPS
 
Pose Estimation (SP)
99% 1930
2.25 IPS
 
Pose Estimation (HP)
99% 1930
2.25 IPS
 
Pose Estimation (Q)
0
 
Object Detection (SP)
99% 263
20.8 IPS
 
Object Detection (HP)
99% 262
20.8 IPS
 
Object Detection (Q)
84% 621
50.1 IPS
 
Face Detection (SP)
100% 888
10.5 IPS
 
Face Detection (HP)
100% 890
10.6 IPS
 
Face Detection (Q)
97% 3417
40.7 IPS
 
Depth Estimation (SP)
83% 1243
9.76 IPS
 
Depth Estimation (HP)
83% 1265
9.93 IPS
 
Depth Estimation (Q)
63% 4984
46.2 IPS
 
Style Transfer (SP)
52% 1951
4.51 IPS
 
Style Transfer (HP)
52% 1941
4.49 IPS
 
Style Transfer (Q)
0
 
Image Super-Resolution (SP)
100% 698
25.8 IPS
 
Image Super-Resolution (HP)
100% 698
25.8 IPS
 
Image Super-Resolution (Q)
97% 5429
201.2 IPS
 
Text Classification (SP)
100% 13
18.0 IPS
 
Text Classification (HP)
100% 13
17.9 IPS
 
Text Classification (Q)
91% 184
246.7 IPS
 
Machine Translation (SP)
100% 43
0.75 IPS
 
Machine Translation (HP)
100% 43
0.75 IPS
 
Machine Translation (Q)
0
 

Compare

Set Baseline Geekbench 6 now available