• @brucethemoose@lemmy.world
      link
      fedilink
      English
      11 month ago

      Some of the distillations are trained on top of Qwen 2.5.

      And for some cases, FuseAI (a special merge of several thinking models), Qwen Coder, EVA-Gutenberg Qwen, or some other specialized models do a better job than Deepseek 32B in certain niches.