GGUF versions?
#1
by
mpasila
- opened
Would you make GGUF versions of these since GGML has been unsupported for a while and will not work on latest llama.cpp versions. Also there are finetuned versions available for 2 of these models Villekom/gpt3-finnish-13B-sft-4epoch and Villekom/gpt3-finnish-3B-sft
(also for the 13B model having a Q3_K_S quantization would be nice to have as well and not just Q4 because I'm pretty memory limited at the moment)