Dark GGUF models
227 GGUF models in the Dark family, quantized by 33 uploaders. Sizes run from 248M to 120B parameters. The longest trained context is 4192K tokens.
About the Dark GGUF models
Dark is represented on ggufmodels.com by 227 GGUF models, quantized by 33 uploaders on Hugging Face. The most downloaded are Dark Scarlett 27B v2.0, L3 Dark Planet 8B and L3 Dark Planet 8B. Parameter counts run from 248M to 120B, and the longest trained context window is 4192K tokens.
A Q4_K_M file of a Dark model is between 0.2 to 61.9 GB, which is the quant most people start with. Most files come from mradermacher (134), ReadyArt (33) and DavidAU (13). The quants you will meet most often are Q4_K_M (206), Q5_K_M (201), Q6_K (195) and Q3_K_M (192), where the number is how many models offer that quant.
Frequently asked questions
Which Dark GGUF should I download?
The most downloaded Dark GGUF here is Dark Scarlett 27B v2.0. 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 Dark model need?
Q4_K_M files of Dark models on this site range from 0.2 to 61.9 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 Dark GGUF files come from?
The most common quantizers here are mradermacher, ReadyArt and DavidAU. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.