CoolFace
Modelpublic

mradermacher/giant-macaroni-120b-i1-GGUF

sourceHugging Facellama2updated 2y agoView on Hugging Face
0likes1.6kdownloads
Model Card

About

weighted/imatrix quants of https://huggingface.co/ibivibiv/giant-macaroni-120b

<!-- provided-files --> static quants are available at https://huggingface.co/mradermacher/giant-macaroni-120b-GGUF

Usage

If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.

Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

LinkTypeSize/GBNotes
GGUFi1-IQ1_S26.2for the desperate
GGUFi1-IQ2_XXS32.8
GGUFi1-IQ2_XS36.3
GGUFi1-IQ2_S38.1
GGUFi1-IQ2_M41.4
GGUFi1-Q2_K45.1IQ3_XXS probably better
GGUFi1-IQ3_XXS47.1lower quality
PART 1 PART 2i1-IQ3_XS50.1
PART 1 PART 2i1-Q3KS52.7IQ3_XS probably better
PART 1 PART 2i1-IQ3_S52.9beats Q3_K*
PART 1 PART 2i1-IQ3_M54.7
PART 1 PART 2i1-Q3KM58.8IQ3_S probably better
PART 1 PART 2i1-Q3KL63.9IQ3_M probably better
PART 1 PART 2i1-IQ4_XS65.1
PART 1 PART 2i1-IQ4_NL68.8prefer IQ4_XS
PART 1 PART 2i1-Q4_068.9fast, low quality
PART 1 PART 2i1-Q4KS69.2optimal size/speed/quality
PART 1 PART 2i1-Q4KM73.1fast, recommended
PART 1 PART 2i1-Q5KS83.7
PART 1 PART 2i1-Q5KM86.0
PART 1 PART 2 PART 3i1-Q6_K99.6practically like static Q6_K

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

image.png

And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

FAQ / Model Request

See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.

Thanks

I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.

<!-- end -->