Geekbench ML Score

601
TensorFlow Lite NNAPI Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date February 11 2025 10:30 PM
Views 3

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 NNAPI
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 282
52.8 IPS
 
Image Classification (F16)
100% 268
50.1 IPS
 
Image Classification (I8)
97% 700
130.9 IPS
 
Image Segmentation (F32)
100% 329
5.50 IPS
 
Image Segmentation (F16)
100% 329
5.50 IPS
 
Image Segmentation (I8)
98% 692
11.6 IPS
 
Pose Estimation (F32)
100% 458
0.55 IPS
 
Pose Estimation (F16)
100% 454
0.55 IPS
 
Pose Estimation (I8)
100% 1920
2.33 IPS
 
Object Detection (F32)
100% 291
21.7 IPS
 
Object Detection (F16)
100% 296
22.1 IPS
 
Object Detection (I8)
61% 807
60.3 IPS
 
Face Detection (F32)
100% 658
7.83 IPS
 
Face Detection (F16)
100% 647
7.69 IPS
 
Face Detection (I8)
86% 1440
17.1 IPS
 
Depth Estimation (F32)
100% 571
4.43 IPS
 
Depth Estimation (F16)
100% 577
4.48 IPS
 
Depth Estimation (I8)
95% 1683
13.0 IPS
 
Style Transfer (F32)
100% 895
1.18 IPS
 
Style Transfer (F16)
100% 903
1.19 IPS
 
Style Transfer (I8)
98% 2273
2.99 IPS
 
Image Super-Resolution (F32)
100% 330
11.8 IPS
 
Image Super-Resolution (F16)
100% 325
11.6 IPS
 
Image Super-Resolution (I8)
98% 1166
41.6 IPS
 
Text Classification (F32)
100% 408
586.1 IPS
 
Text Classification (F16)
100% 416
597.5 IPS
 
Text Classification (I8)
92% 601
863.6 IPS
 
Machine Translation (F32)
100% 642
11.8 IPS
 
Machine Translation (F16)
100% 613
11.3 IPS
 
Machine Translation (I8)
64% 679
12.5 IPS
 

Compare

Set Baseline Geekbench 6 now available