Geekbench ML Score

383
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.1 for iOS AArch64

Result Information

Upload Date October 14 2021 07:13 AM
Views 1

System Information

System Information
Operating System iOS 14.8
Model iPad Pro (9.7-inch)
Model IDiPad6,3
Motherboard J127AP
CPU Information
Name Apple A9X
Topology 1 Processor, 2 Cores
Identifier ARM
Base Frequency 2.26 GHz
L1 Instruction Cache64.0 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache3.00 MB x 1
Memory Information
Size 1.92 GB
Inference Information
Framework TensorFlow Lite
Backend CPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 230
24.7 IPS
 
Image Classification (F16)
100% 232
24.9 IPS
 
Image Classification (I8)
81% 407
43.6 IPS
 
Image Segmentation (F32)
100% 443
2.74 IPS
 
Image Segmentation (F16)
100% 444
2.75 IPS
 
Image Segmentation (I8)
89% 760
4.70 IPS
 
Pose Estimation (F32)
100% 151
0.61 IPS
 
Pose Estimation (F16)
100% 152
0.61 IPS
 
Pose Estimation (I8)
80% 150
0.61 IPS
 
Object Detection (F32)
100% 392
13.7 IPS
 
Object Detection (F16)
100% 389
13.7 IPS
 
Object Detection (I8)
52% 578
20.3 IPS
 
Face Detection (F32)
100% 472
5.43 IPS
 
Face Detection (F16)
100% 473
5.44 IPS
 
Face Detection (I8)
75% 572
6.57 IPS
 
Text Classification (F32)
100% 607
240.2 IPS
 
Text Classification (F16)
100% 605
239.4 IPS
 
Machine Translation (F32)
100% 508
78.3 IPS
 
Machine Translation (F16)
100% 515
79.4 IPS
 

Compare

Set Baseline

Share

Geekbench ML now available for download