Is it just memory bandwidth? Or is it that AMD is not well supported by pytorch well enough for most products? Or some combination of those?

  • Kerfuffle@sh.itjust.worksEnglish
    2·
    10 months ago

    If you’re using llama.cpp, some ROCM stuff recently got merged in. It works pretty well, at least on my 6600. I believe there were instructions for getting it working on Windows in the pull.