Geekbench ML Score

2004
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.2 for iOS AArch64

Result Information

Upload Date January 25 2023 10:33 AM
Views 1

System Information

System Information
Operating System iOS 16.3
Model iPad13,18
Model IDiPad13,18
Motherboard J271AP
CPU Information
Name ARM
Topology 1 Processor, 6 Cores
Identifier ARM
Base Frequency 2.99 GHz
L1 Instruction Cache128 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache4.00 MB x 1
Memory Information
Size 3.74 GB
Inference Information
Framework TensorFlow Lite
Backend GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 2341
250.9 IPS
 
Image Classification (F16)
100% 3984
426.8 IPS
 
Image Classification (I8)
81% 643
68.9 IPS
 
Image Segmentation (F32)
100% 4405
27.2 IPS
 
Image Segmentation (F16)
100% 7457
46.1 IPS
 
Image Segmentation (I8)
89% 1173
7.25 IPS
 
Pose Estimation (F32)
100% 1391
5.63 IPS
 
Pose Estimation (F16)
100% 4551
18.4 IPS
 
Pose Estimation (I8)
80% 223
0.90 IPS
 
Object Detection (F32)
100% 3009
105.6 IPS
 
Object Detection (F16)
99% 3495
122.7 IPS
 
Object Detection (I8)
52% 919
32.2 IPS
 
Face Detection (F32)
100% 3783
43.5 IPS
 
Face Detection (F16)
97% 7563
86.9 IPS
 
Face Detection (I8)
75% 898
10.3 IPS
 
Text Classification (F32)
100% 1538
608.1 IPS
 
Text Classification (F16)
98% 1612
637.7 IPS
 
Machine Translation (F32)
100% 1651
254.4 IPS
 
Machine Translation (F16)
100% 1646
253.7 IPS
 

Compare

Set Baseline