Update status for Qwen3.5 122B and 27B GGUFs?
Hi Unsloth team,
The documentation at "https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks" mentions that 122B and 27B GGUFs are still converting and will be updated soon.
It's been a few days and the model files haven't updated yet. Could you provide an ETA on when these will be ready?
Thanks!
Totally understandable. When you say, "BUT it's iteration 2 for now", what is the latest iteration? 3? No rush, no pressure; just curious.
anyone could quickly do the Aider polyglot benchmark on all of the quants? just start with bf16 and go downwards. thanks! :)
I see Qwen3.5-27B-BF16-00001-of-00002.gguf was updated a day ago, but part 2 still has an old date of 7 days ago. Is it supposed to be this way?
We've announced the updates: https://unsloth.ai/docs/models/qwen3.5
And new benchmarks for qwen3.5 122b
CC: @elpirater312 @McG-221 @asandeistefan @dzupin @Poro7 @oPnf4fMoKMz4VFq @galaxy3856 @mksystem @Orcazephyr @kongarden @electrolanche @tnuvkeg @Simon716 @klimekop6 @wolframko @newsletter @bolt012 @eleius @lightenup @spanspek @irha @Systox @theluminal @cpuQ @gkubon @deraal @AekDevDev @doc-acula @blazingM @nevril
@danielhanchen A sincere and heartfelt "Thank you!", from an MLX user who's noticed the quality of your quants! ❤️🔥