llm
llm copied to clipboard
Support for GGML BLAS
It would be great if BLAS could be enabled in GGML by the user. Enabling BLAS does significantly enhance the performance of GGML. I have tested llama.cpp/GGML with cublas, and it really pays off enabling it.
Thanks, can CBLAS be enabled as well ?
CLBlast can be enabled for Linux and Windows using the PR #282.
Is this done now?
From my POV, it is.