Qwen3.8-Flash-Next
#2969
by jacek2024 - opened
Could you make quants for:
https://huggingface.co/Qwen/Qwen3.8-Flash-Next
(using current llama.cpp with MTP support)
valid GGUFs are here https://huggingface.co/ggml-org/Qwen3.8-Flash-Next-GGUF but these are only Q8 and Q4
It's 1 month old and the code is new
confused, what changed? our thingy doesnt have the mtp? you can use mtp from the repo you provided, it should work fine, unless there was a really major rework that changed something big we dont want to waste a bunch of resources. we are not big company sadly, just a bunch of guys with a couple computer in such expensive times =(
I think you are right, only MTP changed, closing
jacek2024 changed discussion status to closed