realme RMX3708 vs realme RMX3708
| realme RMX3708 | realme RMX3708 | Difference | |
|---|---|---|---|
| Single Precision | 1051 | 1051 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Half Precision | 1022 | 1022 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Quantized | 2564 | 2564 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Geekbench AI 1.1.0 | Geekbench AI 1.1.0 | ||
System Information
| realme RMX3708 | realme RMX3708 | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | CPU | CPU |
| AI Device | ARM ARMv8 | ARM ARMv8 |
| Operating System | Android 14 | Android 14 |
| Model | realme RMX3708 | realme RMX3708 |
| Processor |
ARM ARMv8 @ 1.80 GHz
1 Processor, 8 Cores |
ARM ARMv8 @ 1.80 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 65 architecture 8 variant 2 part 3400 revision 0 | ARM implementer 65 architecture 8 variant 2 part 3400 revision 0 |
| Motherboard | taro | taro |
| Memory | 14.86 GB | 14.86 GB |
AI Performance
| realme RMX3708 | realme RMX3708 | Difference | |
|---|---|---|---|
| Image Classification (SP) | 491 | 491 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Classification (HP) | 491 | 491 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Classification (Q) | 1967 | 1967 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Segmentation (SP) | 588 | 588 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Segmentation (HP) | 592 | 592 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Segmentation (Q) | 2166 | 2166 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Pose Estimation (SP) | 2145 | 2145 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Pose Estimation (HP) | 2154 | 2154 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Pose Estimation (Q) | 4791 | 4791 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Object Detection (SP) | 445 | 445 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Object Detection (HP) | 435 | 435 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Object Detection (Q) | 1634 | 1634 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Face Detection (SP) | 1576 | 1576 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Face Detection (HP) | 1555 | 1555 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Face Detection (Q) | 4175 | 4175 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Depth Estimation (SP) | 1993 | 1993 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Depth Estimation (HP) | 2038 | 2038 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Depth Estimation (Q) | 3899 | 3899 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Style Transfer (SP) | 3519 | 3519 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Style Transfer (HP) | 3502 | 3502 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Style Transfer (Q) | 10729 | 10729 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Super-Resolution (SP) | 1117 | 1117 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Super-Resolution (HP) | 1172 | 1172 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Image Super-Resolution (Q) | 2950 | 2950 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Text Classification (SP) | 617 | 617 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Text Classification (HP) | 457 | 457 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Text Classification (Q) | 1184 | 1184 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Machine Translation (SP) | 785 | 785 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Machine Translation (HP) | 775 | 775 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
| Machine Translation (Q) | 604 | 604 | 100.0% |
| realme RMX3708 | |||
| realme RMX3708 | |||
