Geekbench ML Score

1114
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.2 for iOS AArch64

Result Information

Upload Date November 25 2021 04:29 PM
Views 1

System Information

System Information
Operating System iOS 15.1
Model iPhone XS
Model IDiPhone11,2
Motherboard D321AP
CPU Information
Name Apple A12 Bionic
Topology 1 Processor, 6 Cores
Identifier ARM
Base Frequency 2.49 GHz
L1 Instruction Cache48.0 KB x 1
L1 Data Cache32.0 KB x 1
L2 Cache2.00 MB x 1
Memory Information
Size 3.69 GB
Inference Information
Framework TensorFlow Lite
Backend GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 628
67.2 IPS
 
Image Classification (F16)
100% 1105
118.4 IPS
 
Image Classification (I8)
81% 400
42.9 IPS
 
Image Segmentation (F32)
100% 3061
18.9 IPS
 
Image Segmentation (F16)
100% 4788
29.6 IPS
 
Image Segmentation (I8)
89% 745
4.60 IPS
 
Pose Estimation (F32)
100% 1207
4.88 IPS
 
Pose Estimation (F16)
100% 3450
14.0 IPS
 
Pose Estimation (I8)
80% 141
0.57 IPS
 
Object Detection (F32)
100% 1361
47.8 IPS
 
Object Detection (F16)
99% 1530
53.7 IPS
 
Object Detection (I8)
52% 565
19.8 IPS
 
Face Detection (F32)
100% 3191
36.7 IPS
 
Face Detection (F16)
97% 5035
57.8 IPS
 
Face Detection (I8)
75% 549
6.30 IPS
 
Text Classification (F32)
100% 651
257.4 IPS
 
Text Classification (F16)
98% 798
315.5 IPS
 
Machine Translation (F32)
100% 852
131.4 IPS
 
Machine Translation (F16)
100% 963
148.5 IPS
 

Compare

Set Baseline

Share

Geekbench ML now available for download