Geekbench ML Score

423
TensorFlow Lite CPU Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date July 19 2024 02:21 AM
Views 1

System Information

System Information
Operating System Android 14
Model TECNO TECNO CL7
Model IDTECNO TECNO CL7
Motherboard TECNO-CL7
Governorsugov_ext
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 1 part 3393 revision 0
Base Frequency 2.00 GHz
Cluster 16 Cores @ 2.00 GHz
Cluster 22 Cores @ 2.20 GHz
Memory Information
Size 7.37 GB
Inference Information
Framework TensorFlow Lite
Backend CPU
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 201
37.5 IPS
 
Image Classification (F16)
100% 206
38.5 IPS
 
Image Classification (I8)
97% 408
76.3 IPS
 
Image Segmentation (F32)
100% 288
4.81 IPS
 
Image Segmentation (F16)
100% 289
4.83 IPS
 
Image Segmentation (I8)
98% 484
8.09 IPS
 
Pose Estimation (F32)
100% 405
0.49 IPS
 
Pose Estimation (F16)
100% 404
0.49 IPS
 
Pose Estimation (I8)
100% 1933
2.34 IPS
 
Object Detection (F32)
100% 219
16.3 IPS
 
Object Detection (F16)
100% 217
16.2 IPS
 
Object Detection (I8)
61% 434
32.4 IPS
 
Face Detection (F32)
100% 448
5.32 IPS
 
Face Detection (F16)
100% 446
5.31 IPS
 
Face Detection (I8)
86% 1016
12.1 IPS
 
Depth Estimation (F32)
100% 432
3.35 IPS
 
Depth Estimation (F16)
100% 443
3.44 IPS
 
Depth Estimation (I8)
95% 995
7.72 IPS
 
Style Transfer (F32)
100% 730
0.96 IPS
 
Style Transfer (F16)
100% 744
0.98 IPS
 
Style Transfer (I8)
98% 2089
2.75 IPS
 
Image Super-Resolution (F32)
100% 251
8.95 IPS
 
Image Super-Resolution (F16)
100% 255
9.11 IPS
 
Image Super-Resolution (I8)
98% 657
23.5 IPS
 
Text Classification (F32)
100% 225
322.9 IPS
 
Text Classification (F16)
100% 223
320.9 IPS
 
Text Classification (I8)
92% 207
297.2 IPS
 
Machine Translation (F32)
100% 496
9.13 IPS
 
Machine Translation (F16)
100% 494
9.09 IPS
 
Machine Translation (I8)
64% 306
5.63 IPS
 

Compare

Set Baseline Geekbench 6 now available