• ag10n@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    1 month ago

    You can run it on CPU alone. Not surprising they’re building their own AI ecosystem

    • brucethemoose@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 month ago

      Not at scale. Even on the new architecture, one really needs some kind of accelerator to make it economical for servers.

      Bitnet-like models might change the calculus, but no major trainer had tried that yet.

      • ag10n@lemmy.world
        link
        fedilink
        English
        arrow-up
        0
        ·
        1 month ago

        Yes, you can run it at scale. Which is why it uses Huawei hardware.

        You can run it on anything, scaled or not

        • theunknownmuncher@lemmy.world
          link
          fedilink
          English
          arrow-up
          0
          arrow-down
          1
          ·
          edit-2
          1 month ago

          Nope! You don’t know what you’re talking about. At all. But you can have fun running a 1.6 trillion parameter model on CPU at basically 0 tokens per second at scale, MoE or not.

            • theunknownmuncher@lemmy.world
              link
              fedilink
              English
              arrow-up
              0
              arrow-down
              1
              ·
              edit-2
              1 month ago

              You’ve proved my point that you don’t know what you’re talking about by blindly linking to the git repo. Couldn’t find any source that supports your claim? I wonder why.

              Sure you can serve one request at a time to one patient user at a slow token per second rate, which makes running locally viable, but there is no RAM that has the bandwidth to run this model at scale. Even flash would be incredibly slow on CPU with multiple requests. You’d need the high bandwidth of VRAM and to run across multiple GPUs in a scalable way, it requires extremely high bandwidth interconnects between GPUs.

              • ag10n@lemmy.world
                link
                fedilink
                English
                arrow-up
                0
                ·
                1 month ago

                Thank you for proving my point. It can be run on a cpu

                “It’s slow, it’s inefficient” it still runs

                It’s a foundational model just like R1 was.

                • theunknownmuncher@lemmy.world
                  link
                  fedilink
                  English
                  arrow-up
                  0
                  arrow-down
                  1
                  ·
                  1 month ago

                  Yes, you can run it at scale.

                  at scale

                  Shift those goalposts! We went from “at scale” to “it still runs”

                  • ag10n@lemmy.world
                    link
                    fedilink
                    English
                    arrow-up
                    0
                    ·
                    1 month ago

                    Quote me in full.

                    You can run it at scale, on huawei. You can also run it on a cpu

    • theunknownmuncher@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      arrow-down
      1
      ·
      1 month ago

      I mean, sure. You could also run it by drawing marks in sand. It doesn’t make any sense to do either, though.