Nova GGUF models
90 GGUF models in the Nova family, quantized by 26 uploaders. Sizes run from 30M to 72B parameters. The longest trained context is 256K tokens.
About the Nova GGUF models
Nova is represented on ggufmodels.com by 90 GGUF models, quantized by 26 uploaders on Hugging Face. The most downloaded are Nova-13B, Nova-1.1-0.8B and Nova-v3. Parameter counts run from 30M to 72B, and the longest trained context window is 256K tokens.
A Q4_K_M file of a Nova model is between 0.2 to 45.1 GB, which is the quant most people start with. Most files come from mradermacher (58), VAIBHAV22334455 (3) and silverdollar (2). The quants you will meet most often are Q4_K_M (75), Q5_K_M (65), Q3_K_M (62) and Q2_K (62), where the number is how many models offer that quant.
Frequently asked questions
Which Nova GGUF should I download?
The most downloaded Nova GGUF here is Nova-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 Nova model need?
Q4_K_M files of Nova models on this site range from 0.2 to 45.1 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 Nova GGUF files come from?
The most common quantizers here are mradermacher, VAIBHAV22334455 and silverdollar. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.