Geekbench AI Score

213
Single Precision Score
204
Half Precision Score
472
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

Upload Date June 21 2025 04:18 PM
Views 1

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device vendor
System Information
Operating System Android 12
Model HUAWEI generic_arm_64_anco
Model IDHUAWEI generic_arm_64_anco
Motherboard MRT
CPU Information
Name vendor
Topology 1 Processor, 12 Cores
Identifier ARM implementer 72 architecture 8 variant 2 part 3333 revision 1
Base Frequency 1.31 GHz
Cluster 14 Cores
Cluster 24 Cores @ 1.31 GHz
Cluster 33 Cores @ 2.05 GHz
Cluster 41 Core @ 2.29 GHz
Memory Information
Size 11.38 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 188
35.0 IPS
 
Image Classification (HP)
100% 121
22.6 IPS
 
Image Classification (Q)
99% 310
57.7 IPS
 
Image Segmentation (SP)
100% 161
2.60 IPS
 
Image Segmentation (HP)
100% 162
2.63 IPS
 
Image Segmentation (Q)
98% 335
5.44 IPS
 
Pose Estimation (SP)
100% 207
0.24 IPS
 
Pose Estimation (HP)
100% 207
0.24 IPS
 
Pose Estimation (Q)
98% 781
0.91 IPS
 
Object Detection (SP)
100% 119
9.42 IPS
 
Object Detection (HP)
100% 119
9.42 IPS
 
Object Detection (Q)
87% 336
27.0 IPS
 
Face Detection (SP)
100% 329
3.91 IPS
 
Face Detection (HP)
100% 332
3.95 IPS
 
Face Detection (Q)
97% 722
8.61 IPS
 
Depth Estimation (SP)
100% 266
2.05 IPS
 
Depth Estimation (HP)
99% 268
2.06 IPS
 
Depth Estimation (Q)
64% 647
5.87 IPS
 
Style Transfer (SP)
100% 467
0.60 IPS
 
Style Transfer (HP)
100% 468
0.60 IPS
 
Style Transfer (Q)
98% 1302
1.68 IPS
 
Image Super-Resolution (SP)
100% 128
4.72 IPS
 
Image Super-Resolution (HP)
100% 128
4.74 IPS
 
Image Super-Resolution (Q)
97% 466
17.3 IPS
 
Text Classification (SP)
100% 174
232.3 IPS
 
Text Classification (HP)
100% 174
232.5 IPS
 
Text Classification (Q)
91% 301
405.2 IPS
 
Machine Translation (SP)
100% 292
5.03 IPS
 
Machine Translation (HP)
100% 290
4.99 IPS
 
Machine Translation (Q)
57% 239
5.85 IPS
 

Compare

Set Baseline Geekbench 6 now available