IQ3_M GGUF models
Importance-matrix 3-bit quant, medium. Aimed at low-memory setups. 47889 models have a IQ3_M file. Sizes run from 1M to 1030B parameters. The longest trained context is 102400K tokens.
What is IQ3_M?
Importance-matrix 3-bit quant, medium. Aimed at low-memory setups. On average it stores about 3.7 bits per weight, so a IQ3_M file is roughly 23% of the size of the same model in 16-bit. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
47,889 models have a IQ3_M file on ggufmodels.com, and those files range from 0.0 to 488.1 GB. The most downloaded are Qwen3.8 27B Uncensored HauhauCS Aggressive MTP, Qwen3.5 9B The Defiant Fable Uncensored Heretic NEO IMATRIX MAX MTP and Qwen3.8 27B TURBO Fable Cold Fusion 735 882 Heretic Uncensored NEO CODER MAX MTP. For a longer explanation of every quant type, read GGUF quantization explained.
Frequently asked questions
What is IQ3_M in GGUF?
Importance-matrix 3-bit quant, medium. Aimed at low-memory setups. It averages about 3.7 bits per weight.
Should I use IQ3_M?
A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
How big is a IQ3_M file?
IQ3_M files on this site range from 0.0 to 488.1 GB, depending on the model. As a rule of thumb it is about 23% of the 16-bit size.