Atlas GGUF models
70 GGUF models in the Atlas family, quantized by 26 uploaders. Sizes run from 135M to 72.7B parameters. The longest trained context is 256K tokens.
About the Atlas GGUF models
Atlas is represented on ggufmodels.com by 70 GGUF models, quantized by 26 uploaders on Hugging Face. The most downloaded are N ATLaS, ATLAS 8B Thinking and Atlas Chat 9B. Parameter counts run from 135M to 72.7B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Atlas model is between 0.1 to 47.4 GB, which is the quant most people start with. Most files come from mradermacher (37), ikramknd (4) and RichardErkhov (3). The quants you will meet most often are Q4_K_M (56), Q2_K (46), Q6_K (45) and Q5_K_M (45), where the number is how many models offer that quant.
Frequently asked questions
Which Atlas GGUF should I download?
The most downloaded Atlas GGUF here is N ATLaS. 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 Atlas model need?
Q4_K_M files of Atlas models on this site range from 0.1 to 47.4 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 Atlas GGUF files come from?
The most common quantizers here are mradermacher, ikramknd 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.