Geekbench AI Score

589
Single Precision Score
584
Half Precision Score
1104
Quantized Score
Geekbench AI 1.4.0 for Android AArch64

Result Information

Upload Date June 26 2025 07:11 PM
Views 3

System Information

AI Information
Framework TensorFlow Lite
Backend CPU
Device ARM ARMv8
System Information
Operating System Android 14
Model OnePlus OPD2203
Model IDOnePlus OPD2203
Motherboard k6983v1_64
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 2 part 3400 revision 0
Base Frequency 1.80 GHz
Cluster 14 Cores @ 1.80 GHz
Cluster 23 Cores @ 2.85 GHz
Cluster 31 Core @ 3.05 GHz
Memory Information
Size 7.49 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 282
52.5 IPS
 
Image Classification (HP)
100% 282
52.4 IPS
 
Image Classification (Q)
100% 817
152.0 IPS
 
Image Segmentation (SP)
100% 340
5.51 IPS
 
Image Segmentation (HP)
100% 349
5.67 IPS
 
Image Segmentation (Q)
98% 787
12.8 IPS
 
Pose Estimation (SP)
100% 1264
1.47 IPS
 
Pose Estimation (HP)
100% 1261
1.47 IPS
 
Pose Estimation (Q)
84% 2598
3.08 IPS
 
Object Detection (SP)
98% 303
24.1 IPS
 
Object Detection (HP)
98% 308
24.5 IPS
 
Object Detection (Q)
83% 754
60.9 IPS
 
Face Detection (SP)
100% 715
8.50 IPS
 
Face Detection (HP)
100% 701
8.33 IPS
 
Face Detection (Q)
95% 1770
21.1 IPS
 
Depth Estimation (SP)
99% 960
7.42 IPS
 
Depth Estimation (HP)
99% 979
7.56 IPS
 
Depth Estimation (Q)
64% 1850
16.9 IPS
 
Style Transfer (SP)
89% 2857
3.71 IPS
 
Style Transfer (HP)
89% 2828
3.67 IPS
 
Style Transfer (Q)
98% 5554
7.16 IPS
 
Image Super-Resolution (SP)
100% 509
18.8 IPS
 
Image Super-Resolution (HP)
100% 519
19.2 IPS
 
Image Super-Resolution (Q)
97% 1246
46.2 IPS
 
Text Classification (SP)
100% 286
382.0 IPS
 
Text Classification (HP)
99% 265
353.1 IPS
 
Text Classification (Q)
88% 498
672.3 IPS
 
Machine Translation (SP)
100% 483
8.32 IPS
 
Machine Translation (HP)
100% 456
7.86 IPS
 
Machine Translation (Q)
50% 190
6.74 IPS
 

Compare

Set Baseline Geekbench 6 now available