About

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.

(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	15.8
GGUF	Q3_K_S	18.3
GGUF	Q3_K_M	20.1	lower quality
GGUF	Q3_K_L	21.7
GGUF	IQ4_XS	22.4
GGUF	Q4_K_S	23.5	fast, recommended
GGUF	Q4_K_M	24.7	fast, recommended
GGUF	Q5_K_S	28.2
GGUF	Q5_K_M	28.9
GGUF	Q6_K	33.3	very good quality
GGUF	Q8_0	43.1	fast, best quality

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

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.