samsung SM-T978U vs samsung SM-T978U
| samsung SM-T978U | samsung SM-T978U | Difference | |
|---|---|---|---|
| Single Precision | 698 | 698 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Half Precision | 624 | 624 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Quantized | 1045 | 1045 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Geekbench AI 1.2.0 | Geekbench AI 1.2.0 | ||
System Information
| samsung SM-T978U | samsung SM-T978U | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | CPU | CPU |
| AI Device | ARM Qualcomm | ARM Qualcomm |
| Operating System | Android 12 | Android 12 |
| Model | samsung SM-T978U | samsung SM-T978U |
| Processor |
ARM Qualcomm @ 1.80 GHz
1 Processor, 8 Cores |
ARM Qualcomm @ 1.80 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 65 architecture 8 variant 1 part 3341 revision 0 | ARM implementer 65 architecture 8 variant 1 part 3341 revision 0 |
| Motherboard | kona | kona |
| Memory | 5.50 GB | 5.50 GB |
AI Performance
| samsung SM-T978U | samsung SM-T978U | Difference | |
|---|---|---|---|
| Image Classification (SP) | 172 | 172 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Classification (HP) | 178 | 178 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Classification (Q) | 1444 | 1444 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Segmentation (SP) | 581 | 581 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Segmentation (HP) | 580 | 580 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Segmentation (Q) | 1614 | 1614 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Pose Estimation (SP) | 1977 | 1977 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Pose Estimation (HP) | 1843 | 1843 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Pose Estimation (Q) | 3177 | 3177 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Object Detection (SP) | 390 | 390 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Object Detection (HP) | 154 | 154 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Object Detection (Q) | 414 | 414 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Face Detection (SP) | 1412 | 1412 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Face Detection (HP) | 1434 | 1434 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Face Detection (Q) | 3086 | 3086 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Depth Estimation (SP) | 1759 | 1759 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Depth Estimation (HP) | 1418 | 1418 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Depth Estimation (Q) | 1006 | 1006 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Style Transfer (SP) | 3413 | 3413 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Style Transfer (HP) | 3492 | 3492 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Style Transfer (Q) | 6263 | 6263 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Super-Resolution (SP) | 979 | 979 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Super-Resolution (HP) | 1002 | 1002 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Image Super-Resolution (Q) | 639 | 639 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Text Classification (SP) | 164 | 164 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Text Classification (HP) | 165 | 165 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Text Classification (Q) | 252 | 252 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Machine Translation (SP) | 263 | 263 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Machine Translation (HP) | 262 | 262 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
| Machine Translation (Q) | 162 | 162 | 100.0% |
| samsung SM-T978U | |||
| samsung SM-T978U | |||
