Link	Type	Size/GB	Notes
GGUF	i1-IQ1_S	26.4	for the desperate
GGUF	i1-IQ1_M	28.9	mostly desperate
GGUF	i1-IQ2_XXS	33.2
GGUF	i1-IQ2_XS	36.8
GGUF	i1-IQ2_S	38.6
GGUF	i1-IQ2_M	42.0
GGUF	i1-Q2_K	45.8	IQ3_XXS probably better
GGUF	i1-IQ3_XXS	47.9	lower quality
PART 1 PART 2	i1-IQ3_XS	51.0
PART 1 PART 2	i1-Q3_K_S	53.7	IQ3_XS probably better
PART 1 PART 2	i1-IQ3_S	53.9	beats Q3_K*
PART 1 PART 2	i1-IQ3_M	55.7
PART 1 PART 2	i1-Q3_K_M	59.9	IQ3_S probably better
PART 1 PART 2	i1-Q3_K_L	65.2	IQ3_M probably better
PART 1 PART 2	i1-IQ4_XS	66.4
PART 1 PART 2	i1-Q4_0	70.4	fast, low quality
PART 1 PART 2	i1-Q4_K_S	70.6	optimal size/speed/quality
PART 1 PART 2	i1-Q4_K_M	74.6	fast, recommended
PART 1 PART 2	i1-Q5_K_S	85.5
PART 1 PART 2	i1-Q5_K_M	87.8
PART 1 PART 2 PART 3	i1-Q6_K	101.9	practically like static Q6_K

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

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.