Google Pixel 3a
Geekbench AI Score
97
Single Precision Score
156
Half Precision Score
152
Quantized Score
Geekbench AI 1.0.0 for Android AArch64
Result Information
| User | grinceur |
| Upload Date | August 23 2024 03:30 PM |
| Views | 8 |
System Information
| AI Information | |
|---|---|
| Framework | TensorFlow Lite |
| Backend | GPU |
| Device | Qualcomm Snapdragon 670 |
| System Information | |
|---|---|
| Operating System | Android 12 |
| Model | Google Pixel 3a |
| Model ID | Google Pixel 3a |
| Motherboard | sargo |
| Governor | schedutil |
| CPU Information | |
|---|---|
| Name | Qualcomm Qualcomm |
| Topology | 1 Processor, 8 Cores |
| Identifier | ARM implementer 81 architecture 8 variant 6 part 2050 revision 13 |
| Base Frequency | 1.71 GHz |
| Cluster 1 | 6 Cores @ 1.71 GHz |
| Cluster 2 | 2 Cores @ 2.00 GHz |
| Memory Information | |
|---|---|
| Size | 3.51 GB |
AI Performance
| Workload | Accuracy | Score | |
|---|---|---|---|
|
Image Classification (SP)
|
100% |
61
11.4 IPS |
|
|
Image Classification (HP)
|
100% |
103
19.2 IPS |
|
|
Image Classification (Q)
|
97% |
109
20.3 IPS |
|
|
Image Segmentation (SP)
|
100% |
83
1.35 IPS |
|
|
Image Segmentation (HP)
|
100% |
152
2.47 IPS |
|
|
Image Segmentation (Q)
|
98% |
151
2.46 IPS |
|
|
Pose Estimation (SP)
|
100% |
117
0.14 IPS |
|
|
Pose Estimation (HP)
|
100% |
185
0.22 IPS |
|
|
Pose Estimation (Q)
|
95% |
184
0.22 IPS |
|
|
Object Detection (SP)
|
100% |
55
4.37 IPS |
|
|
Object Detection (HP)
|
100% |
89
7.02 IPS |
|
|
Object Detection (Q)
|
86% |
92
7.37 IPS |
|
|
Face Detection (SP)
|
100% |
148
1.76 IPS |
|
|
Face Detection (HP)
|
100% |
271
3.22 IPS |
|
|
Face Detection (Q)
|
97% |
288
3.43 IPS |
|
|
Depth Estimation (SP)
|
100% |
145
1.12 IPS |
|
|
Depth Estimation (HP)
|
99% |
267
2.06 IPS |
|
|
Depth Estimation (Q)
|
75% |
256
2.07 IPS |
|
|
Style Transfer (SP)
|
100% |
277
0.36 IPS |
|
|
Style Transfer (HP)
|
100% |
640
0.82 IPS |
|
|
Style Transfer (Q)
|
98% |
635
0.82 IPS |
|
|
Image Super-Resolution (SP)
|
100% |
72
2.64 IPS |
|
|
Image Super-Resolution (HP)
|
100% |
131
4.84 IPS |
|
|
Image Super-Resolution (Q)
|
97% |
129
4.79 IPS |
|
|
Text Classification (SP)
|
100% |
49
65.1 IPS |
|
|
Text Classification (HP)
|
100% |
49
64.9 IPS |
|
|
Text Classification (Q)
|
91% |
59
79.0 IPS |
|
|
Machine Translation (SP)
|
100% |
115
1.98 IPS |
|
|
Machine Translation (HP)
|
100% |
112
1.93 IPS |
|
|
Machine Translation (Q)
|
57% |
70
1.70 IPS |
