Code GGUF models
132 GGUF models in the Code family, quantized by 42 uploaders. Sizes run from 42M to 33.7B parameters. The longest trained context is 986K tokens.
About the Code GGUF models
Code is represented on ggufmodels.com by 132 GGUF models, quantized by 42 uploaders on Hugging Face. The most downloaded are Code 13B, Code 290k 13B and models--mergekit-community--mergekit-slerp-fecxcxs. Parameter counts run from 42M to 33.7B, and the longest trained context window is 986K tokens.
A Q4_K_M file of a Code model is between 0.1 to 20.2 GB, which is the quant most people start with. Most files come from mradermacher (42), microtensor-archive (11) and RichardErkhov (11). The quants you will meet most often are Q4_K_M (75), Q2_K (67), Q5_K_M (65) and Q3_K_M (65), where the number is how many models offer that quant.
Frequently asked questions
Which Code GGUF should I download?
The most downloaded Code GGUF here is Code 13B. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Code model need?
Q4_K_M files of Code models on this site range from 0.1 to 20.2 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Code GGUF files come from?
The most common quantizers here are mradermacher, microtensor-archive and RichardErkhov. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.