Geekbench AI Score

183
Single Precision Score
273
Half Precision Score
854
Quantized Score
Geekbench AI 1.2.0 for Android AArch64

Result Information

User grinceur
Upload Date February 11 2025 03:41 PM
Views 8

System Information

AI Information
Framework TensorFlow Lite
Backend NNAPI
Device Qualcomm Snapdragon 730G
System Information
Operating System Android 13
Model Google Pixel 4a
Model IDGoogle Pixel 4a
Motherboard sunfish
Governorschedutil
CPU Information
Name Qualcomm Qualcomm
Topology 1 Processor, 8 Cores
Identifier ARM implementer 81 architecture 8 variant 13 part 2052 revision 14
Base Frequency 1.80 GHz
Cluster 16 Cores @ 1.80 GHz
Cluster 22 Cores @ 2.21 GHz
Memory Information
Size 5.47 GB

AI Performance

Workload Accuracy Score
Image Classification (SP)
100% 180
33.4 IPS
 
Image Classification (HP)
100% 253
47.1 IPS
 
Image Classification (Q)
97% 560
104.6 IPS
 
Image Segmentation (SP)
100% 235
3.80 IPS
 
Image Segmentation (HP)
100% 150
2.43 IPS
 
Image Segmentation (Q)
98% 668
10.9 IPS
 
Pose Estimation (SP)
100% 473
0.55 IPS
 
Pose Estimation (HP)
100% 1106
1.29 IPS
 
Pose Estimation (Q)
96% 4102
4.80 IPS
 
Object Detection (SP)
100% 177
14.0 IPS
 
Object Detection (HP)
100% 110
8.69 IPS
 
Object Detection (Q)
85% 617
49.7 IPS
 
Face Detection (SP)
100% 409
4.86 IPS
 
Face Detection (HP)
100% 269
3.19 IPS
 
Face Detection (Q)
97% 1476
17.6 IPS
 
Depth Estimation (SP)
99% 288
2.22 IPS
 
Depth Estimation (HP)
99% 229
1.76 IPS
 
Depth Estimation (Q)
63% 2440
22.7 IPS
 
Style Transfer (SP)
100% 496
0.64 IPS
 
Style Transfer (HP)
100% 377
0.49 IPS
 
Style Transfer (Q)
98% 2288
2.95 IPS
 
Image Super-Resolution (SP)
100% 283
10.5 IPS
 
Image Super-Resolution (HP)
100% 547
20.2 IPS
 
Image Super-Resolution (Q)
97% 1734
64.2 IPS
 
Text Classification (SP)
71% 4
5.59 IPS
 
Text Classification (HP)
100% 155
206.7 IPS
 
Text Classification (Q)
93% 118
158.8 IPS
 
Machine Translation (SP)
100% 188
3.25 IPS
 
Machine Translation (HP)
100% 258
4.44 IPS
 
Machine Translation (Q)
54% 130
3.66 IPS
 

Compare

Set Baseline Geekbench 6 now available