Catch up on all the horror stories from big tech - from May. And some tech, film, blog and book tips.

  • Matt@lemmy.ml
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    2 days ago

    Just buy 10 of these, install Linux and K3s on them, deploy Ollama, some UI frontend, merge the GPUs with https://github.com/gpustack/gpustack and you’ve got yourself a pretty good AI setup that is local and that also has (outside of electricity bills) close to zero maintenance cost.

    • iocase@lemmy.zip
      link
      fedilink
      arrow-up
      1
      ·
      edit-2
      2 days ago

      just spend €11 000 and run 10 entire laptops so you can have your own local Ollama instance