Daniel Holanda

Fresh builds of llama.cpp with AMD ROCm™ 7 acceleration

Enabling Flexible FPGA High-Level Synthesis of Tensorflow Deep Neural Networks