Geekbench ML Score

473
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date May 07 2024 01:32 AM
Views 1

System Information

System Information
Operating System Android 14
Model Xiaomi 22101320G
Model IDXiaomi 22101320G
Motherboard redwood
Governorschedutil
CPU Information
Name ARM redwood
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 1
Base Frequency 1.80 GHz
Cluster 14 Cores @ 1.80 GHz
Cluster 24 Cores @ 2.40 GHz
Memory Information
Size 5.07 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 222
41.6 IPS
 
Image Classification (F16)
100% 227
42.4 IPS
 
Image Classification (I8)
97% 511
95.7 IPS
 
Image Segmentation (F32)
100% 317
5.29 IPS
 
Image Segmentation (F16)
100% 315
5.27 IPS
 
Image Segmentation (I8)
98% 467
7.80 IPS
 
Pose Estimation (F32)
100% 488
0.59 IPS
 
Pose Estimation (F16)
100% 482
0.58 IPS
 
Pose Estimation (I8)
100% 2325
2.82 IPS
 
Object Detection (F32)
100% 240
17.9 IPS
 
Object Detection (F16)
100% 243
18.2 IPS
 
Object Detection (I8)
61% 622
46.4 IPS
 
Face Detection (F32)
100% 471
5.60 IPS
 
Face Detection (F16)
100% 471
5.60 IPS
 
Face Detection (I8)
86% 1193
14.2 IPS
 
Depth Estimation (F32)
100% 480
3.72 IPS
 
Depth Estimation (F16)
100% 490
3.80 IPS
 
Depth Estimation (I8)
95% 1118
8.67 IPS
 
Style Transfer (F32)
100% 845
1.11 IPS
 
Style Transfer (F16)
100% 849
1.12 IPS
 
Style Transfer (I8)
98% 1993
2.62 IPS
 
Image Super-Resolution (F32)
100% 290
10.3 IPS
 
Image Super-Resolution (F16)
100% 273
9.75 IPS
 
Image Super-Resolution (I8)
98% 891
31.8 IPS
 
Text Classification (F32)
100% 259
371.6 IPS
 
Text Classification (F16)
100% 252
361.7 IPS
 
Text Classification (I8)
92% 248
357.0 IPS
 
Machine Translation (F32)
100% 521
9.59 IPS
 
Machine Translation (F16)
100% 556
10.2 IPS
 
Machine Translation (I8)
64% 237
4.37 IPS
 

Compare

Set Baseline Geekbench 6 now available