LM Studio produces gibberish (GGUF)
#1
by jrell - opened
I have downloaded the Q4_K_M file and ran it with LM Studio, but the result is the following:
I tried changing the settings to see if the problem was on my end, but it didn't help. Any ideas?
Ah, my bad. I created a thread under the wrong model. Should I close it?
jrell changed discussion title from LM Studio produces gibberish to LM Studio produces gibberish (GGUF)
I have the same issue with kobold with the Q8 version
Yea I accidentally made a mistake when merging the LORA back to base which caused issues when making the quants. I also do need to reupload this base model because of this. The 8B version seems to work fine though.
Hey everyone, I have fixed the model and also reuploaded all the GGUFs now. It should work properly.
jrell changed discussion status to closed