Base GGUF models
127 GGUF models in the Base family, quantized by 43 uploaders. Sizes run from 37M to 105B parameters. The longest trained context is 1024K tokens.
About the Base GGUF models
Base is represented on ggufmodels.com by 127 GGUF models, quantized by 43 uploaders on Hugging Face. The most downloaded are internistai base 7b v0.2, T5 Base and L2 7b Base Guanaco Uncensored i1. Parameter counts run from 37M to 105B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a Base model is between 0.1 to 42.5 GB, which is the quant most people start with. Most files come from mradermacher (58), xlight05 (6) and tensorblock (6). The quants you will meet most often are Q4_K_M (78), Q2_K (69), Q5_K_M (63) and Q4_K_S (62), where the number is how many models offer that quant.
Frequently asked questions
Which Base GGUF should I download?
The most downloaded Base GGUF here is internistai base 7b v0.2. 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 Base model need?
Q4_K_M files of Base models on this site range from 0.1 to 42.5 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 Base GGUF files come from?
The most common quantizers here are mradermacher, xlight05 and tensorblock. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.