Anbernic RG477M vs Anbernic RG477M
| Anbernic RG477M | Anbernic RG477M | Difference | |
|---|---|---|---|
| Single Precision | 1143 | 1143 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Half Precision | 1514 | 1514 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Quantized | 1260 | 1260 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Geekbench AI 1.5.0 | Geekbench AI 1.5.0 | ||
System Information
| Anbernic RG477M | Anbernic RG477M | |
|---|---|---|
| AI Framework | TensorFlow Lite | TensorFlow Lite |
| AI Backend | GPU | GPU |
| AI Device | ARM ARMv8 | ARM ARMv8 |
| Operating System | Android 14 | Android 14 |
| Model | Anbernic RG477M | Anbernic RG477M |
| Processor |
ARM ARMv8 @ 2.20 GHz
1 Processor, 8 Cores |
ARM ARMv8 @ 2.20 GHz
1 Processor, 8 Cores |
| Processor ID | ARM implementer 65 architecture 8 variant 1 part 3405 revision 0 | ARM implementer 65 architecture 8 variant 1 part 3405 revision 0 |
| Motherboard | k6897v1_64 | k6897v1_64 |
| Memory | 7.64 GB | 7.64 GB |
AI Performance
| Anbernic RG477M | Anbernic RG477M | Difference | |
|---|---|---|---|
| Image Classification (SP) | 648 | 648 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Classification (HP) | 656 | 656 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Classification (Q) | 581 | 581 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Segmentation (SP) | 1507 | 1507 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Segmentation (HP) | 2784 | 2784 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Segmentation (Q) | 2203 | 2203 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Pose Estimation (SP) | 7242 | 7242 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Pose Estimation (HP) | 11272 | 11272 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Pose Estimation (Q) | 12180 | 12180 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Object Detection (SP) | 574 | 574 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Object Detection (HP) | 615 | 615 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Object Detection (Q) | 753 | 753 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Face Detection (SP) | 2379 | 2379 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Face Detection (HP) | 3814 | 3814 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Face Detection (Q) | 2763 | 2763 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Depth Estimation (SP) | 3230 | 3230 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Depth Estimation (HP) | 5115 | 5115 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Depth Estimation (Q) | 4029 | 4029 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Style Transfer (SP) | 10114 | 10114 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Style Transfer (HP) | 18562 | 18562 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Style Transfer (Q) | 20761 | 20761 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Super-Resolution (SP) | 2026 | 2026 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Super-Resolution (HP) | 3085 | 3085 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Image Super-Resolution (Q) | 2727 | 2727 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Text Classification (SP) | 17 | 17 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Text Classification (HP) | 11 | 11 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Text Classification (Q) | 14 | 14 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Machine Translation (SP) | 353 | 353 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Machine Translation (HP) | 408 | 408 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
| Machine Translation (Q) | 98 | 98 | 100.0% |
| Anbernic RG477M | |||
| Anbernic RG477M | |||
