• doodledup@lemmy.world
    link
    fedilink
    English
    arrow-up
    28
    arrow-down
    6
    ·
    4 months ago

    Alexa and LLMs are fundamentally not too different from each other. It’s just a slightly different architecture and most importantly a much larger network.

    The problem with LLMs is that they require immense compute power.

    I don’t see how LLMs will get into the households any time soon. It’s not economical.

    • just another dev@lemmy.my-box.dev
      link
      fedilink
      English
      arrow-up
      15
      ·
      4 months ago

      The problem with LLMs is that they require immense compute power.

      To train. But you can run a relatively simple one like phi-3 on quite modest hardware.

    • hedgehog@ttrpg.network
      link
      fedilink
      English
      arrow-up
      5
      arrow-down
      1
      ·
      4 months ago

      I don’t see how LLMs will get into the households any time soon. It’s not economical.

      I can run an LLM on my phone, on my tablet, on my laptop, on my desktop, or on my server. Heck, I could run a small model on the Raspberry PI 5 if I wanted. And none of those devices have dedicated chips for AI.

      The problem with LLMs is that they require immense compute power.

      Not really, particularly if you’re talking about the usage of smaller models. Running an LLM on your GPU and sending it queries isn’t going to use more energy than using your GPU to game for the same amount of time would.

      • doodledup@lemmy.world
        link
        fedilink
        English
        arrow-up
        3
        ·
        4 months ago

        I think when people talk about LLMs replacing Alexa they mean the much more capable models with billions of parameters. The small models that a Raspberry-Pi can run are no use really.

        • hedgehog@ttrpg.network
          link
          fedilink
          English
          arrow-up
          5
          ·
          4 months ago

          The models I’m talking about that a PI 5 can run have billions of parameters, though. For example, Mistral 7B (here’s a guide to running it on the PI 5) has roughly 7 Billion parameters. By quantizing each parameter to 4 bits, it only takes up 3.5 GB in RAM, making it easily fit in the 8 GB model’s memory. If you have a GPU with 8+ GB of VRAM (most cards from the past few years have 8 GB or more - the 1070, 2060 Super, and 3050 and each better card in that generation hit that mark), you have enough VRAM and more than enough speed to run Q4 versions of the 13B models (which have roughly 13 Billion parameters), and if you have one with 24 GB of VRAM, like the 3090, then you can run Q4 versions of the 30B models.

          Apple Silicon Macs can also competently run inference for these models - for them, the limiting factor is system RAM, not VRAM, though. And it’s not like you’ll need a Mac as even Microsoft is investing in ARM CPUs with dedicated AI chips.

          • doodledup@lemmy.world
            link
            fedilink
            English
            arrow-up
            2
            ·
            4 months ago

            Thanks for sharing that. I have a Raspberry-Pi 4B laying around and getting dusty. I might try this.

      • doodledup@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        4 months ago

        Well yea. You could slap Gemini Google-Home today. You wouldn’t even need a new device for that probably. The reason they don’t do that is econimical.

        My point is that LLMs aren’t replacing those devices. They are the same thing essentially. Just one a trimmed version of the other for economic reasons.

    • Halcyon@discuss.tchncs.de
      link
      fedilink
      English
      arrow-up
      1
      ·
      4 months ago

      The immense computing power for AI is needed for training LLMs, it’s far less for running a pre-trained model on a local machine.