Llama.cpp Now Part of the Nvidia RTX AI Toolkit(developer.nvidia.com)
developer.nvidia.com
Llama.cpp Now Part of the Nvidia RTX AI Toolkit
https://developer.nvidia.com/blog/accelerating-llms-with-llama-cpp-on-nvidia-rtx-systems/
https://developer.nvidia.com/blog/accelerating-llms-with-llama-cpp-on-nvidia-rtx-systems/
I don't want ROCM to be the only way to use HIP.
Just upstream stuff, I don't want some special proprietary package.