Sao10K/Fimbulvetr-11B-v2
Model name:
Fimbulvetr-11B-v2
Model link:
https://huggingface.co/Sao10K/Fimbulvetr-11B-v2
Brief description:
One of, if not the best, RP models under 70B parameters I've ever used. While I know that you don't quantize anything past 7B/8B, I feel Fimbulvetr 11B v2 has been such a well-loved and utilized model that it deserves a proper iMatrix quant by the best RP model quantization provider I know—you. (Yes, I know, I am sucking up a bit, but the point still stands. The other imatrix quants of this model just don't feel well done enough...) A Q6 would be greatly appreciated.
An image/direct image link to represent the model:
https://huggingface.co/Sao10K/Fimbulvetr-11B-v2/resolve/main/cute1.jpg
While I know that you don't quantize anything past 7B/8B, I feel Fimbulvetr 11B v2 has been such a well-loved and utilized model [...].
I do prefer doing the smaller sizes but the 11B should be doable. Will let you know when it's ready.
The other imatrix quants of this model just don't feel well done enough...
Not sure but hey, I'll do the best possible here (imatrix data was generated from the FP16-GGUF and conversions as well since the original model weights are FP16), hopefully it's not something wrong with the model itself.
Cheers! <3
@Ayalexf
Alright! Currently uploading everything at:
https://huggingface.co/Lewdiculous/Fimbulvetr-11B-v2-GGUF-IQ-Imatrix
@Ayalexf Alright! Currently uploading everything at:
https://huggingface.co/Lewdiculous/Fimbulvetr-11B-v2-GGUF-IQ-Imatrix
God bless!