Geekbench ML Score

2522
TensorFlow Lite GPU Inference Score
Geekbench ML 0.5.2 for iOS AArch64

Result Information

Upload Date September 23 2022 05:40 AM
Views 1

System Information

System Information
Operating System iOS 16.1
Model iPhone15,3
Model IDiPhone15,3
Motherboard D74AP
CPU Information
Name ARM
Topology 1 Processor, 6 Cores
Identifier ARM
Base Frequency 3.44 GHz
L1 Instruction Cache128 KB x 1
L1 Data Cache64.0 KB x 1
L2 Cache4.00 MB x 1
Memory Information
Size 5.49 GB
Inference Information
Framework TensorFlow Lite
Backend GPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 2913
312.1 IPS
 
Image Classification (F16)
100% 4568
489.4 IPS
 
Image Classification (I8)
81% 744
79.7 IPS
 
Image Segmentation (F32)
100% 6398
39.5 IPS
 
Image Segmentation (F16)
100% 9335
57.7 IPS
 
Image Segmentation (I8)
89% 1365
8.43 IPS
 
Pose Estimation (F32)
100% 2384
9.64 IPS
 
Pose Estimation (F16)
100% 5136
20.8 IPS
 
Pose Estimation (I8)
80% 258
1.04 IPS
 
Object Detection (F32)
100% 3362
118.0 IPS
 
Object Detection (F16)
100% 3739
131.2 IPS
 
Object Detection (I8)
52% 1066
37.4 IPS
 
Face Detection (F32)
100% 5876
67.5 IPS
 
Face Detection (F16)
97% 9966
114.5 IPS
 
Face Detection (I8)
75% 1041
12.0 IPS
 
Text Classification (F32)
100% 2143
847.4 IPS
 
Text Classification (F16)
100% 1873
740.9 IPS
 
Machine Translation (F32)
100% 2273
350.3 IPS
 
Machine Translation (F16)
100% 2275
350.7 IPS
 

Compare

Set Baseline

Share

Geekbench ML now available for download