• BB84@mander.xyzOP
    link
    fedilink
    English
    arrow-up
    2
    ·
    edit-2
    12 days ago

    They said they’re working on Orthus for Qwen 3.5. It’ll be amazing!

    • tristynalxander@mander.xyz
      link
      fedilink
      English
      arrow-up
      6
      ·
      edit-2
      11 days ago

      Yeah, unfortunately it seems this can’t be converted to a llama.cpp compatible format yet, and that’s pretty big a tradeoff right now. Not surprising with how new it is, but we’ll have to wait to combine it with other improvements. Pretty exciting for the future though.

      Update: I actually couldn’t get this to run even on HuggingFace Transformers. I made a bug report, but basically I’m getting some torch incompatibilities with flash-attn. Maybe this is a known issue for more experienced folks, but I couldn’t solve it.