What is that?

• Standard Q4_K_M and Q4_K_S quants of DarkArtsForge/Morbid-Miasma-24B
• BF16 weights hosted temporarily for easier enthusiast quantization
• imatrix file

Quant Details

• vanilla Q4_K_S and Q4_K_M

• quantized with ik_llama.cpp build: 9d07d868

• imatrix generated using bartowski's calibration_datav5.txt.

Cheers

MistralAI - the beloved base model(s).
ikawrakow and contributors of ik_llama.cpp - I probably misused your creation.
DarkArtsForge/Naphula - for the effort of designing and producing the merge. And for reviving my interest in Mistral 24b.
+++ Everyone whose finetunes were included in the merge!
bartowski - for the calibration data + the myriad of quants we all benefit from.

Disclosure

imatrix generated on an intermediary q8_0 quant. I don't eat RAM for breakfast.
My only contribution is compute. This is neither my merge, nor my calibration data. Have fun.