Question?
#1
by
Alkohole
- opened
It's a curious mix. Will you be releasing the Q8 and below?
Currently i do not have enough space for that, but anyone can do the quants. Just use the b16 and throw it into lcpp and quantize it. And the only reason i uploaded the gguf is because someone asked me to do so because they wanted to test it locally. I could upload the q4_k_s as that is what i used for testing.
Edited: I went and uploaded the Q4_k_s anyways.
Added all the quants up to iq4_xs. as anything below that isn't worth using for this kind of size.
Alkohole
changed discussion status to
closed