Anbernic RGCUBE vs Anbernic RGCUBE
| Anbernic RGCUBE | Anbernic RGCUBE | Difference | |
|---|---|---|---|
| Single Precision | 689 | 689 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Half Precision | 766 | 766 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Quantized | 1627 | 1627 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Geekbench AI 1.4.0 | Geekbench AI 1.4.0 | ||
System Information
| Anbernic RGCUBE | Anbernic RGCUBE | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | CPU | CPU |
| AI Device | ARM Unisoc | ARM Unisoc |
| Operating System | Android 13 | Android 13 |
| Model | Anbernic RGCUBE | Anbernic RGCUBE |
| Processor |
ARM Unisoc @ 2.18 GHz
1 Processor, 8 Cores |
ARM Unisoc @ 2.18 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 65 architecture 8 variant 4 part 3339 revision 0 | ARM implementer 65 architecture 8 variant 4 part 3339 revision 0 |
| Motherboard | ums9620_2h10 | ums9620_2h10 |
| Memory | 7.71 GB | 7.71 GB |
AI Performance
| Anbernic RGCUBE | Anbernic RGCUBE | Difference | |
|---|---|---|---|
| Image Classification (SP) | 419 | 419 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Classification (HP) | 436 | 436 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Classification (Q) | 1183 | 1183 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Segmentation (SP) | 444 | 444 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Segmentation (HP) | 447 | 447 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Segmentation (Q) | 1368 | 1368 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Pose Estimation (SP) | 1508 | 1508 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Pose Estimation (HP) | 1551 | 1551 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Pose Estimation (Q) | 2993 | 2993 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Object Detection (SP) | 402 | 402 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Object Detection (HP) | 407 | 407 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Object Detection (Q) | 1156 | 1156 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Face Detection (SP) | 887 | 887 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Face Detection (HP) | 891 | 891 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Face Detection (Q) | 2726 | 2726 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Depth Estimation (SP) | 1118 | 1118 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Depth Estimation (HP) | 1122 | 1122 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Depth Estimation (Q) | 2823 | 2823 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Style Transfer (SP) | 1064 | 1064 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Style Transfer (HP) | 2873 | 2873 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Style Transfer (Q) | 6707 | 6707 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Super-Resolution (SP) | 678 | 678 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Super-Resolution (HP) | 674 | 674 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Image Super-Resolution (Q) | 1685 | 1685 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Text Classification (SP) | 463 | 463 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Text Classification (HP) | 458 | 458 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Text Classification (Q) | 726 | 726 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Machine Translation (SP) | 649 | 649 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Machine Translation (HP) | 644 | 644 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
| Machine Translation (Q) | 369 | 369 | 100.0% |
| Anbernic RGCUBE | |||
| Anbernic RGCUBE | |||
