VAETKI-GGUF

GGUF format of VAETKI for use with llama.cpp.

Model Overview

VAETKI is a large language model developed by the NC-AI consortium, a collaborative initiative led by NC-AI with participation from a total of 13 organizations.

• Type: Causal (Auto-regressive) Language Models
• Architecture: Transformers, MoE (Mixture of Experts)
• Developed by: NC-AI consortium (with ETRI, Korea University)
• Number of Parameters: 112.2B in total and 10.1B activated
• Number of Layers: 48
• Number of Attention Heads: 24
• Number of Experts: 128
• Number of Activated Experts: 8
• Context Length: 32k tokens
• Vocabulary Size: 126k
• Languages: Korean, English, Chinese, and Japanese
• License: MIT

Available Files

Filename	Quant	Size
VAETKI-F16.gguf	F16	~209GB
VAETKI-Q4_K_M.gguf	Q4_K_M	~63GB

Usage

Note: VAETKI support requires a modified version of llama.cpp. Until the PR is merged, please use: https://github.com/dororodoroddo/llama.cpp/tree/add-vaetki-support

./llama-cli -m VAETKI-Q4_K_M.gguf -p "Your prompt here"

License

MIT License. See the original model repository for details.

Citation

@misc{ncai2025vaetkitechnicalreport,
      title={VAETKI Technical Report},
      author={NC-AI Consortium},
      year={2025},
      eprint={xxxx.xxxxx},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/xxxx.xxxxx},
}

Contact

wbl.ncai.hf@gmail.com