Geekbench ML Score

1529
TensorFlow Lite NNAPI Inference Score
Geekbench ML 0.6.0 for Android AArch64

Result Information

Upload Date September 03 2025 04:10 AM
Views 1

System Information

System Information
Operating System Android 14
Model alps k6878v1_64_ssdev
Model IDalps k6878v1_64_ssdev
Motherboard k6878v1_64
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 14 Cores @ 2.00 GHz
Cluster 24 Cores @ 2.50 GHz
Memory Information
Size 7.20 GB
Inference Information
Framework TensorFlow Lite
Backend NNAPI
Device Default

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 429
80.2 IPS
 
Image Classification (F16)
100% 1510
282.5 IPS
 
Image Classification (I8)
97% 4539
849.4 IPS
 
Image Segmentation (F32)
100% 548
9.16 IPS
 
Image Segmentation (F16)
100% 1563
26.1 IPS
 
Image Segmentation (I8)
98% 1990
33.2 IPS
 
Pose Estimation (F32)
100% 2652
3.21 IPS
 
Pose Estimation (F16)
100% 5923
7.17 IPS
 
Pose Estimation (I8)
100% 26094
31.6 IPS
 
Object Detection (F32)
100% 312
23.3 IPS
 
Object Detection (F16)
100% 971
72.5 IPS
 
Object Detection (I8)
64% 1782
133.1 IPS
 
Face Detection (F32)
100% 1316
15.7 IPS
 
Face Detection (F16)
97% 3750
44.6 IPS
 
Face Detection (I8)
86% 8750
104.0 IPS
 
Depth Estimation (F32)
100% 401
3.11 IPS
 
Depth Estimation (F16)
100% 6931
53.7 IPS
 
Depth Estimation (I8)
95% 14301
110.9 IPS
 
Style Transfer (F32)
100% 1644
2.16 IPS
 
Style Transfer (F16)
100% 1469
1.93 IPS
 
Style Transfer (I8)
31% 4392
5.78 IPS
 
Image Super-Resolution (F32)
100% 968
34.6 IPS
 
Image Super-Resolution (F16)
100% 4255
151.9 IPS
 
Image Super-Resolution (I8)
33% 10641
380.0 IPS
 
Text Classification (F32)
100% 68
97.9 IPS
 
Text Classification (F16)
100% 652
936.5 IPS
 
Text Classification (I8)
92% 859
1.23 KIPS
 
Machine Translation (F32)
100% 184
3.39 IPS
 
Machine Translation (F16)
100% 179
3.30 IPS
 
Machine Translation (I8)
66% 312
5.74 IPS
 

Compare

Set Baseline Geekbench 6 now available