Geekbench ML Score

3351
TensorFlow Lite NNAPI Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date September 20 2024 12:13 PM
Views 3

System Information

System Information
Operating System Android 14
Model vivo PA2473
Model IDvivo PA2473
Motherboard tb8796p1_64_sp
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 0 part 3458 revision 1
Base Frequency 2.00 GHz
Cluster 14 Cores @ 2.00 GHz
Cluster 23 Cores @ 2.85 GHz
Cluster 31 Core @ 3.25 GHz
Memory Information
Size 15.20 GB
Inference Information
Framework TensorFlow Lite
Backend NNAPI
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 970
181.6 IPS
 
Image Classification (F16)
100% 4507
843.3 IPS
 
Image Classification (I8)
97% 8264
1.55 KIPS
 
Image Segmentation (F32)
100% 649
10.8 IPS
 
Image Segmentation (F16)
100% 4061
67.8 IPS
 
Image Segmentation (I8)
98% 1790
29.9 IPS
 
Pose Estimation (F32)
100% 8249
9.99 IPS
 
Pose Estimation (F16)
100% 17793
21.5 IPS
 
Pose Estimation (I8)
100% 66820
80.9 IPS
 
Object Detection (F32)
100% 692
51.7 IPS
 
Object Detection (F16)
100% 2913
217.4 IPS
 
Object Detection (I8)
64% 3288
245.5 IPS
 
Face Detection (F32)
100% 3392
40.3 IPS
 
Face Detection (F16)
97% 12214
145.2 IPS
 
Face Detection (I8)
86% 11481
136.5 IPS
 
Depth Estimation (F32)
100% 930
7.22 IPS
 
Depth Estimation (F16)
100% 12624
97.9 IPS
 
Depth Estimation (I8)
95% 21979
170.4 IPS
 
Style Transfer (F32)
100% 2553
3.36 IPS
 
Style Transfer (F16)
100% 2391
3.14 IPS
 
Style Transfer (I8)
31% 4693
6.17 IPS
 
Image Super-Resolution (F32)
100% 2122
75.8 IPS
 
Image Super-Resolution (F16)
100% 11275
402.7 IPS
 
Image Super-Resolution (I8)
33% 33509
1.20 KIPS
 
Text Classification (F32)
100% 124
177.9 IPS
 
Text Classification (F16)
100% 1712
2.46 KIPS
 
Text Classification (I8)
76% 1808
2.60 KIPS
 
Machine Translation (F32)
100% 391
7.20 IPS
 
Machine Translation (F16)
100% 2196
40.4 IPS
 
Machine Translation (I8)
76% 553
10.2 IPS
 

Compare

Set Baseline Geekbench 6 now available