Originally published at: NVDLA Deep Learning Inference Compiler is Now Open Source | NVIDIA Technical Blog
Designing new custom hardware accelerators for deep learning is clearly popular, but achieving state-of-the-art performance and efficiency with a new design is a complex and challenging problem. Two years ago, NVIDIA opened the source for the hardware design of the NVIDIA Deep Learning Accelerator (NVDLA) to help advance the adoption of efficient AI inferencing in…
and yet the NVIDIA's driver stack and CUDA compilers are not open source. I advise anyone who cares about software Freedom to avoid CUDA and do not use NVIDIA hardware until they stop their monopolistic policies.