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

base_model	datasets	language	library_name	license	quantized_by	tags
Isotonic/smol_llama-4x220M-MoE	JeanKaddour/minipile pszemraj/simple_wikipedia_LM mattymchen/refinedweb-3m HuggingFaceH4/ultrachat_200k teknium/openhermes HuggingFaceH4/ultrafeedback_binarized EleutherAI/proof-pile-2 bigcode/the-stack-smol-xl	en	transformers	apache-2.0	mradermacher	moe merge mergekit lazymergekit BEE-spoke-data/smol_llama-220M-openhermes BEE-spoke-data/beecoder-220M-python BEE-spoke-data/zephyr-220m-sft-full BEE-spoke-data/zephyr-220m-dpo-full text-generation

About

static quants of https://huggingface.co/Isotonic/smol_llama-4x220M-MoE

weighted/imatrix quants are available at https://huggingface.co/mradermacher/smol_llama-4x220M-MoE-i1-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	Q2_K	0.3
GGUF	Q3_K_S	0.4
GGUF	Q3_K_M	0.4	lower quality
GGUF	Q3_K_L	0.4
GGUF	IQ4_XS	0.4
GGUF	Q4_K_S	0.4	fast, recommended
GGUF	Q4_K_M	0.5	fast, recommended
GGUF	Q5_K_S	0.5
GGUF	Q5_K_M	0.5
GGUF	Q6_K	0.6	very good quality
GGUF	Q8_0	0.7	fast, best quality
GGUF	f16	1.3	16 bpw, overkill

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.