Geekbench ML Score

3433
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.2 for iOS AArch64

Result Information

Upload Date December 08 2022 06:19 PM
Views 1

System Information

System Information
Operating System iOS 16.1.1
Model iPad14,3
Model IDiPad14,3
Motherboard J617AP
CPU Information
Name ARM
Topology 1 Processor, 8 Cores
Identifier ARM
Base Frequency 3.47 GHz
L1 Instruction Cache128 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache4.00 MB x 1
Memory Information
Size 7.35 GB
Inference Information
Framework TensorFlow Lite
Backend GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 4700
503.6 IPS
 
Image Classification (F16)
100% 7007
750.8 IPS
 
Image Classification (I8)
81% 754
80.8 IPS
 
Image Segmentation (F32)
100% 11099
68.6 IPS
 
Image Segmentation (F16)
100% 16742
103.4 IPS
 
Image Segmentation (I8)
89% 1376
8.50 IPS
 
Pose Estimation (F32)
100% 3831
15.5 IPS
 
Pose Estimation (F16)
100% 9306
37.7 IPS
 
Pose Estimation (I8)
80% 260
1.05 IPS
 
Object Detection (F32)
100% 5264
184.7 IPS
 
Object Detection (F16)
100% 5509
193.3 IPS
 
Object Detection (I8)
52% 1077
37.8 IPS
 
Face Detection (F32)
100% 9464
108.7 IPS
 
Face Detection (F16)
97% 15863
182.3 IPS
 
Face Detection (I8)
75% 1039
11.9 IPS
 
Text Classification (F32)
100% 3612
1.43 KIPS
 
Text Classification (F16)
100% 3966
1.57 KIPS
 
Machine Translation (F32)
100% 1923
296.4 IPS
 
Machine Translation (F16)
100% 1917
295.4 IPS
 

Compare

Set Baseline