Unsloth Dynamic 3.0 GGUFs

28 points | by jonesy827 an hour ago

4 comments

  • throwa356262 4 minutes ago

       "We also made some smaller UD-1bit quants with UD-IQ1_S being 6.2GB (without MTP) which retain around 72% top-1% accuracy yet being 89% smaller"
    
    
    This is crazy! But has anyone tried these lower quants on real projects?
  • xlayn 8 minutes ago

    Hey Unsloth, your gguf are the first ones I look for when I want to download a gguf model. Today I was trying in fact to see, what's the smallest Qwen3.8-27B that I could run and get good results, say restricting it to 16GB of ram.. so I went, pick up the Qwen3.8-27B-UD-IQ2_XXS.gguf and them BAM, error on MTP... now I understand why after reading your announcement. Beyond the space saving, why removing the MTP? improves speed exactly for the group that could benefit from it.

      mike-the-brain 7 minutes ago

      you can still have it, no?

      > We also removed the MTP module from smaller quants under UD-Q2_K_XL (8.37GB and lower) to converse around 500MB of disk space - you can use the Q4_0 MTP separate module if needed

  • mike-the-brain 8 minutes ago

    Might be off-topic but: is it possible to perform such a quantization on Apple devices? Something like Mac Studio Ultra M1 (even if it would take weeks/months)?