• crispyflagstones@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    1
    ·
    edit-2
    6 months ago

    The completely different software stack is a killer. It’s not that you can’t find versions of a model to run, but almost everything that hits the GPU for compute is going to be targeting CUDA, not RocM. From a compatibility standpoint alone this killed AMD for me. I just do not want to spend my time fighting the stack to get these models running.

    • WormFood@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      6 months ago

      on the one hand, cuda is vendor lock-in and if we’d all just agreed on an open standard decades ago then we wouldn’t be in this mess

      but on the other hand, rocm is crap and adaptivecpp is very half baked right now, at least in my limited experience

      • crispyflagstones@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        2
        ·
        edit-2
        6 months ago

        Yeah, it’s not that I like this state of affairs, but right now the vendor lock-in is so one-sided that it’s hard to say there’s a viable alternative to CUDA. I hope that changes one day.

    • aard@kyu.de
      link
      fedilink
      English
      arrow-up
      1
      ·
      6 months ago

      Admittedly I’m just toying around for entertainment purposes - but I didn’t really have any problems of getting anything I wanted to try out with rocm support. Bigger annoyance was different projects targetting specific distributions or specific software versions (mostly ancient python), but as I’m doing everything in containers anyway that also was manageable.