Geekbench AI Score

213
Single Precision Score
213
Half Precision Score
490
Quantized Score
Geekbench AI 1.1.0 for Android AArch64

Result Information

Upload Date September 24 2024 08:37 AM
Views 4

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device ARM ARMv8
System Information
Operating System Android 14
Model vivo I2217
Model IDvivo I2217
Motherboard taro
Governorwalt
CPU Information
Name ARM ARMv8
Topology 1 Processor, 8 Cores
Identifier ARM implementer 65 architecture 8 variant 2 part 3400 revision 0
Base Frequency 1.80 GHz
Cluster 14 Cores @ 1.80 GHz
Cluster 23 Cores @ 2.50 GHz
Cluster 31 Core @ 3.00 GHz
Memory Information
Size 11.06 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 135
25.1 IPS
 
Image Classification (HP)
100% 136
25.3 IPS
 
Image Classification (Q)
99% 360
67.1 IPS
 
Image Segmentation (SP)
100% 179
2.90 IPS
 
Image Segmentation (HP)
100% 180
2.92 IPS
 
Image Segmentation (Q)
98% 390
6.35 IPS
 
Pose Estimation (SP)
100% 242
0.28 IPS
 
Pose Estimation (HP)
100% 234
0.27 IPS
 
Pose Estimation (Q)
98% 814
0.95 IPS
 
Object Detection (SP)
100% 128
10.1 IPS
 
Object Detection (HP)
100% 128
10.2 IPS
 
Object Detection (Q)
87% 336
27.0 IPS
 
Face Detection (SP)
100% 314
3.74 IPS
 
Face Detection (HP)
100% 316
3.75 IPS
 
Face Detection (Q)
97% 686
8.18 IPS
 
Depth Estimation (SP)
100% 268
2.07 IPS
 
Depth Estimation (HP)
99% 269
2.08 IPS
 
Depth Estimation (Q)
64% 648
5.88 IPS
 
Style Transfer (SP)
100% 482
0.62 IPS
 
Style Transfer (HP)
100% 482
0.62 IPS
 
Style Transfer (Q)
98% 1288
1.66 IPS
 
Image Super-Resolution (SP)
100% 133
4.92 IPS
 
Image Super-Resolution (HP)
100% 133
4.91 IPS
 
Image Super-Resolution (Q)
97% 465
17.2 IPS
 
Text Classification (SP)
100% 170
227.2 IPS
 
Text Classification (HP)
100% 171
228.1 IPS
 
Text Classification (Q)
91% 302
405.6 IPS
 
Machine Translation (SP)
100% 290
5.00 IPS
 
Machine Translation (HP)
100% 289
4.98 IPS
 
Machine Translation (Q)
57% 262
6.41 IPS
 

Compare

Set Baseline Geekbench 6 now available