• Alphane Moon@lemmy.worldOPM
    link
    fedilink
    English
    arrow-up
    6
    ·
    20 days ago

    It’s not too bad to have the capability to run a local LLM or have your dGPU accelerate video upscaling algorithms, but “GiMate AI chatbot” is definitely not it.

    • nate3d@lemmy.world
      link
      fedilink
      English
      arrow-up
      2
      ·
      20 days ago

      Oh absolutely. I’ve got two Nvidia Jetsons and honestly they’re perfect for running dedicated workloads or automating training runs. I’d love for MB manufacturers to start tossing a soldered NPU/TPU on the board and call it a day