Geekbench ML Score

954
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date July 15 2024 12:57 PM
Views 1

System Information

System Information
Operating System Android 14
Model Samsung Galaxy S23 Ultra
Model IDsamsung SM-S918B
Motherboard kalama
Governorwalt
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3406 revision 0
Base Frequency 2.02 GHz
Cluster 13 Cores @ 2.02 GHz
Cluster 24 Cores @ 2.80 GHz
Cluster 31 Core @ 3.36 GHz
Memory Information
Size 10.77 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 455
85.2 IPS
 
Image Classification (F16)
100% 478
89.4 IPS
 
Image Classification (I8)
97% 862
161.3 IPS
 
Image Segmentation (F32)
100% 601
10.0 IPS
 
Image Segmentation (F16)
100% 594
9.91 IPS
 
Image Segmentation (I8)
98% 713
11.9 IPS
 
Pose Estimation (F32)
100% 1126
1.36 IPS
 
Pose Estimation (F16)
100% 1057
1.28 IPS
 
Pose Estimation (I8)
100% 4403
5.33 IPS
 
Object Detection (F32)
100% 486
36.3 IPS
 
Object Detection (F16)
100% 482
36.0 IPS
 
Object Detection (I8)
61% 1100
82.2 IPS
 
Face Detection (F32)
100% 1037
12.3 IPS
 
Face Detection (F16)
100% 1050
12.5 IPS
 
Face Detection (I8)
86% 1784
21.2 IPS
 
Depth Estimation (F32)
100% 1037
8.04 IPS
 
Depth Estimation (F16)
100% 1041
8.07 IPS
 
Depth Estimation (I8)
95% 1965
15.2 IPS
 
Style Transfer (F32)
100% 1790
2.35 IPS
 
Style Transfer (F16)
100% 1790
2.35 IPS
 
Style Transfer (I8)
98% 4087
5.38 IPS
 
Image Super-Resolution (F32)
100% 629
22.5 IPS
 
Image Super-Resolution (F16)
100% 687
24.5 IPS
 
Image Super-Resolution (I8)
98% 1833
65.5 IPS
 
Text Classification (F32)
100% 623
895.5 IPS
 
Text Classification (F16)
100% 614
882.9 IPS
 
Text Classification (I8)
92% 466
670.2 IPS
 
Machine Translation (F32)
100% 1019
18.7 IPS
 
Machine Translation (F16)
100% 1008
18.5 IPS
 
Machine Translation (I8)
64% 487
8.96 IPS
 

Compare

Set Baseline Geekbench 6 now available