Based on the AI-specific APiM framework, a modular deep neural network learning accelerator without any external caching can be used for high-performance edge computing, as a vision-based deep learning computing and AI algorithm acceleration.
Firefly-RK3399/AIO-3399J Android8.1 firmware supports Android Neural Networks API (NNAPI), which fully invokes the neural network API for hardware acceleration, and improves the AI computing performance of the RK3399 greatly.
The kernel code of Firefly ROC-RK3328-CC has been officially approved by kernel.org. The related BSP support has been added to Linux Kernel 4.17 version.
Firefly first ultra-small open-source board ROC-RK3328-CC, using the RK3328 quad-core A53x4 64-bit processor, the size as a credit card, but contains a strong performance.