Geekbench ML Score

596
TensorFlow Lite GPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date November 10 2024 01:29 PM
Views 1

System Information

System Information
Operating System Android 12
Model Xiaomi Redmi K50 Pro
Model IDXiaomi 22011211C
Motherboard matisse
Governorsugov_ext
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 2 part 3400 revision 0
Base Frequency 1.60 GHz
Cluster 14 Cores @ 1.60 GHz
Cluster 23 Cores @ 2.85 GHz
Cluster 31 Core @ 3.00 GHz
Memory Information
Size 11.19 GB
Inference Information
Framework TensorFlow Lite
Backend GPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 384
71.8 IPS
 
Image Classification (F16)
100% 493
92.3 IPS
 
Image Classification (I8)
97% 405
75.7 IPS
 
Image Segmentation (F32)
100% 413
6.90 IPS
 
Image Segmentation (F16)
100% 312
5.21 IPS
 
Image Segmentation (I8)
98% 371
6.20 IPS
 
Pose Estimation (F32)
100% 553
0.67 IPS
 
Pose Estimation (F16)
100% 789
0.96 IPS
 
Pose Estimation (I8)
100% 788
0.95 IPS
 
Object Detection (F32)
100% 230
17.2 IPS
 
Object Detection (F16)
100% 221
16.5 IPS
 
Object Detection (I8)
62% 228
17.0 IPS
 
Face Detection (F32)
100% 1382
16.4 IPS
 
Face Detection (F16)
98% 1367
16.3 IPS
 
Face Detection (I8)
86% 1545
18.4 IPS
 
Depth Estimation (F32)
100% 812
6.29 IPS
 
Depth Estimation (F16)
100% 996
7.73 IPS
 
Depth Estimation (I8)
95% 1006
7.80 IPS
 
Style Transfer (F32)
100% 2143
2.82 IPS
 
Style Transfer (F16)
100% 3093
4.07 IPS
 
Style Transfer (I8)
98% 3110
4.09 IPS
 
Image Super-Resolution (F32)
100% 440
15.7 IPS
 
Image Super-Resolution (F16)
100% 626
22.4 IPS
 
Image Super-Resolution (I8)
98% 591
21.1 IPS
 
Text Classification (F32)
100% 234
335.7 IPS
 
Text Classification (F16)
100% 237
340.9 IPS
 
Text Classification (I8)
92% 480
690.4 IPS
 
Machine Translation (F32)
100% 363
6.67 IPS
 
Machine Translation (F16)
100% 369
6.78 IPS
 
Machine Translation (I8)
64% 450
8.29 IPS
 

Compare

Set Baseline Geekbench 6 now available