GGUF models by Naphula
92 GGUF models quantized by Naphula on Hugging Face. Sizes run from 3.88B to 107B parameters. The longest trained context is 1048K tokens.
About Naphula on Hugging Face
Naphula has published 92 GGUF models that ggufmodels.com tracks, across families such as Naphula, Mistral and Llama. The most downloaded are KrakenSakura Maelstrom 12B v1, Slimaki Tavern 24B v1.3 and Naphula__Slimaki 24B v1.
The quants Naphula publishes most often are Q6_K (81), Q4_K_M (31), Q8_K_XL (21) and Q5_K_M (21). Their Q4_K_M files range from 4.4 to 43.9 GB. Parameter counts run from 3.88B to 107B, and the longest trained context window is 1048K tokens.
Frequently asked questions
What GGUF models has Naphula published?
92 models are listed here, mainly from the Naphula, Mistral and Llama families. The most downloaded is KrakenSakura Maelstrom 12B v1.
Which quants does Naphula publish?
The most common are Q6_K (81 models), Q4_K_M (31 models), Q8_K_XL (21 models) and Q5_K_M (21 models).