Q3_K_L GGUF models
3-bit k-quant, large. A little better than Q3_K_M at a slightly larger size. 101200 models have a Q3_K_L file. Sizes run from 1M to 1030B parameters. The longest trained context is 4194304K tokens.
What is Q3_K_L?
3-bit k-quant, large. A little better than Q3_K_M at a slightly larger size. On average it stores about 4.3 bits per weight, so a Q3_K_L file is roughly 27% 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.
101,200 models have a Q3_K_L file on ggufmodels.com, and those files range from 0.0 to 530.5 GB. The most downloaded are Qwen3.8-27B, Ornith 1.5 9B Uncensored and Qwen3.8 Flash Next Uncensored. For a longer explanation of every quant type, read GGUF quantization explained.
Frequently asked questions
What is Q3_K_L in GGUF?
3-bit k-quant, large. A little better than Q3_K_M at a slightly larger size. It averages about 4.3 bits per weight.
Should I use Q3_K_L?
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 Q3_K_L file?
Q3_K_L files on this site range from 0.0 to 530.5 GB, depending on the model. As a rule of thumb it is about 27% of the 16-bit size.