Basically the title.

For me, a dense model around 20b, with an additional engram table of 10-20b, or a sparse MoE around 80b would be amazing.

  • ZephyrXero@lemmy.world
    link
    fedilink
    English
    arrow-up
    4
    ·
    1 month ago

    They’ve changed the license with Qwen 4, starting with 3.8-next. So I probably won’t be using any further releases

    • fluxx@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      ·
      1 month ago

      What are the significant changes? Requiring attribution or something more sinister?