realme RMX5108 vs realme RMX5108
| realme RMX5108 | realme RMX5108 | Difference | |
|---|---|---|---|
| Single Precision | 538 | 538 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Half Precision | 819 | 819 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Quantized | 626 | 626 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Geekbench AI 1.5.0 | Geekbench AI 1.5.0 | ||
System Information
| realme RMX5108 | realme RMX5108 | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | GPU | GPU |
| AI Device | ARM ARMv8 | ARM ARMv8 |
| Operating System | Android 15 | Android 15 |
| Model | realme RMX5108 | realme RMX5108 |
| Processor |
ARM ARMv8 @ 2.00 GHz
1 Processor, 8 Cores |
ARM ARMv8 @ 2.00 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 65 architecture 8 variant 1 part 3393 revision 0 | ARM implementer 65 architecture 8 variant 1 part 3393 revision 0 |
| Motherboard | RM6878 | RM6878 |
| Memory | 7.29 GB | 7.29 GB |
AI Performance
| realme RMX5108 | realme RMX5108 | Difference | |
|---|---|---|---|
| Image Classification (SP) | 269 | 269 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Classification (HP) | 440 | 440 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Classification (Q) | 394 | 394 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Segmentation (SP) | 562 | 562 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Segmentation (HP) | 991 | 991 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Segmentation (Q) | 969 | 969 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Pose Estimation (SP) | 2511 | 2511 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Pose Estimation (HP) | 4530 | 4530 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Pose Estimation (Q) | 4390 | 4390 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Object Detection (SP) | 257 | 257 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Object Detection (HP) | 427 | 427 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Object Detection (Q) | 369 | 369 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Face Detection (SP) | 1102 | 1102 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Face Detection (HP) | 1632 | 1632 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Face Detection (Q) | 1409 | 1409 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Depth Estimation (SP) | 1704 | 1704 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Depth Estimation (HP) | 2423 | 2423 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Depth Estimation (Q) | 2174 | 2174 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Style Transfer (SP) | 4277 | 4277 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Style Transfer (HP) | 7664 | 7664 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Style Transfer (Q) | 7579 | 7579 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Super-Resolution (SP) | 716 | 716 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Super-Resolution (HP) | 1056 | 1056 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Image Super-Resolution (Q) | 1060 | 1060 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Text Classification (SP) | 12 | 12 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Text Classification (HP) | 14 | 14 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Text Classification (Q) | 7 | 7 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Machine Translation (SP) | 306 | 306 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Machine Translation (HP) | 360 | 360 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
| Machine Translation (Q) | 88 | 88 | 100.0% |
| realme RMX5108 | |||
| realme RMX5108 | |||
