TheBloke's quants?
#5
by stolsvik - opened
Hi @TheBloke - would it be possible to quantize this? Would be interesting to try.
Maybe this problem? https://github.com/ggerganov/llama.cpp/issues/4331
Hi @TheBloke - would it be possible to quantize this? Would be interesting to try.
Maybe this problem? https://github.com/ggerganov/llama.cpp/issues/4331