base_model, datasets, language, library_name, license, mradermacher, quantized_by, tags

base_model	datasets	language	library_name	license	mradermacher	quantized_by	tags
shisa-ai/ablation-195-finalsft2-shisa-v2-mistral-small-24b	shisa-ai/shisa-v2-best-of-n-athenev2-tulu70b-llama33-only-no-sysprompt shisa-ai/shisa-v2-roleplaying-sft shisa-ai/translation_set_april_6 shisa-ai/rewild-set-deepseek-subset shisa-ai/magpie-ultra-set shisa-ai/magpie-advanced-questions-set shisa-ai/japan-magpie-set shisa-ai/shisa-v2-instruction-following-sft	en	transformers	apache-2.0	readme_rev 1	mradermacher	generated_from_trainer

About

weighted/imatrix quants of https://huggingface.co/shisa-ai/ablation-195-finalsft2-shisa-v2-mistral-small-24b

For a convenient overview and download list, visit our model page for this model.

static quants are available at https://huggingface.co/mradermacher/ablation-195-finalsft2-shisa-v2-mistral-small-24b-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)

Link	Type	Size/GB	Notes
GGUF	i1-IQ1_S	5.4	for the desperate
GGUF	i1-IQ1_M	5.9	mostly desperate
GGUF	i1-IQ2_XXS	6.6
GGUF	i1-IQ2_XS	7.3
GGUF	i1-IQ2_S	7.6
GGUF	i1-IQ2_M	8.2
GGUF	i1-Q2_K_S	8.4	very low quality
GGUF	i1-Q2_K	9.0	IQ3_XXS probably better
GGUF	i1-IQ3_XXS	9.4	lower quality
GGUF	i1-IQ3_XS	10.0
GGUF	i1-Q3_K_S	10.5	IQ3_XS probably better
GGUF	i1-IQ3_S	10.5	beats Q3_K*
GGUF	i1-IQ3_M	10.8
GGUF	i1-Q3_K_M	11.6	IQ3_S probably better
GGUF	i1-Q3_K_L	12.5	IQ3_M probably better
GGUF	i1-IQ4_XS	12.9
GGUF	i1-Q4_0	13.6	fast, low quality
GGUF	i1-Q4_K_S	13.6	optimal size/speed/quality
GGUF	i1-Q4_K_M	14.4	fast, recommended
GGUF	i1-Q4_1	15.0
GGUF	i1-Q5_K_S	16.4
GGUF	i1-Q5_K_M	16.9
GGUF	i1-Q6_K	19.4	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

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. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.