Modelpublic

mradermacher/ShiningValiantXS-i1-GGUF

sourceHugging Facellama2updated 2y agoView on Hugging Face

0likes4kdownloads

Model Card

weighted/imatrix quants of https://huggingface.co/ValiantLabs/Llama2-13B-ShiningValiant

static quants are available at https://huggingface.co/mradermacher/ShiningValiantXS-GGUF

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.

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

Link	Type	Size/GB	Notes
GGUF	i1-IQ1_S	3.5	for the desperate
GGUF	i1-IQ2_XXS	4.2
GGUF	i1-IQ2_XS	4.5
GGUF	i1-IQ2_S	4.8
GGUF	i1-IQ2_M	5.2
GGUF	i1-Q2_K	5.5	IQ3_XXS probably better
GGUF	i1-IQ3_XXS	5.6	lower quality
GGUF	i1-IQ3_XS	6.0
GGUF	i1-IQ3_S	6.3	beats Q3_K*
GGUF	i1-Q3KS	6.3	IQ3_XS probably better
GGUF	i1-IQ3_M	6.6
GGUF	i1-Q3KM	7.0	IQ3_S probably better
GGUF	i1-Q3KL	7.6	IQ3_M probably better
GGUF	i1-IQ4_XS	7.6
GGUF	i1-Q4KS	8.0	optimal size/speed/quality
GGUF	i1-Q4KM	8.5	fast, recommended
GGUF	i1-Q5KS	9.6
GGUF	i1-Q5KM	9.9
GGUF	i1-Q6_K	11.3	practically like static Q6_K

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

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

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

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.