text_encoders/qwen_omni_bf16.safetensors 8.1 GBdiffusion_models/auk_base_fp32.safetensors 6.1 GBdiffusion_models/auk_flash_fp32.safetensors 6.1 GBtext_encoders/qwen_omni_int8.safetensors 4.7 GBtext_encoders/qwen_omni_w4a8.safetensors 3.2 GBdiffusion_models/auk_base_bf16.safetensors 3.1 GBdiffusion_models/auk_flash_bf16.safetensors 3.1 GBdiffusion_models/auk_base_int8.safetensors 1.5 GBdiffusion_models/auk_flash_int8.safetensors 1.5 GBdiffusion_models/auk_base_w4a8.safetensors 880 MBdiffusion_models/auk_flash_w4a8.safetensors 880 MBvae/auk_vae.safetensors 637 MBLocal speech generation, voice cloning and editing with AuK (Base & Flash) in ComfyUI. 8 nodes, fully local, BF16 / INT8 / W4A8 / FP32 checkpoints.
| Folder | Files |
|---|---|
diffusion_models/ |
auk_base / auk_flash × fp32 / bf16 / int8 / w4a8 |
text_encoders/ |
qwen_omni_bf16 · qwen_omni_int8 · qwen_omni_w4a8 |
vae/ |
auk_vae (unquantized) |
Drop files directly into the matching ComfyUI/models/ folders — only weights needed, config/tokenizer ship with the node pack.
⚠️ Prompt adherence varies by task (voice cloning, TTS and some edits are reliable; others need retries). This is a model limitation — see the tested task ranking before downloading.
Research using AuK can cite: arXiv:2609.08936
@misc{ma2026auktechnicalreportopensource,
title={AuK Technical Report: An Open-Source Foundational Model for Speech Generation and Editing},
author={Ziyang Ma and others},
year={2026},
eprint={2609.08936},
archivePrefix={arXiv},
primaryClass={cs.SD},
url={https://arxiv.org/abs/2609.08936},
}
Have the files? Submit a torrent.
Submit a torrentHave the complete files and permission to share them? Replace the example folder path. The script checks the files, makes a torrent and submits its magnet and checksums. Keep your torrent client seeding. Get your community key from your settings, and read the script before running it.
curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo drbaph/AuK-comfyui ./model-folder