nunchux-ai

license

Apache-2.0(OSI

source

public(nunchux-ai/nunchaku)

core features withheld

no — no ee, enterprise, commercial or proprietary path

Nunchaku is released under Apache 2.0 with its full source public. Its authors offer consulting and partnership to companies but no separate paid build of the engine.

• https://raw.githubusercontent.com/nunchux-ai/nunchaku/main/LICENCE.txt recorded 2026-09-26

  LICENCE.txt body: "Apache License / Version 2.0, January 2004".

• https://raw.githubusercontent.com/nunchux-ai/nunchaku/main/README.md recorded 2026-09-26

  README Contact Us: "For enterprises interested in adopting SVDQuant or Nunchaku, including technical consulting, sponsorship opportunities, or partnership inquiries, please contact us"; no paid edition is named.

• https://ungh.cc/repos/nunchux-ai/nunchaku/files/main recorded 2026-09-26

  Full file list: no ee/, enterprise/, commercial/ or proprietary/ path; LICENCE.txt is the only license file.

The engine is distributed as wheels on its GitHub releases, and the PyPI name it would use belongs to an unrelated project, so GitHub stars are the only comparable signal. A star is not a use.

• https://pypi.org/pypi/nunchaku/json recorded 2026-09-26

  PyPI nunchaku: "Optimally partitioning data into piece-wise linear segments.", with its repository at git.ecdf.ed.ac.uk, an unrelated project.

• https://ungh.cc/repos/nunchux-ai/nunchaku recorded 2026-09-26

  Repository record: stars 3955.

Nunchaku runs several image model families in 4-bit form, fitting large models such as FLUX.1 onto consumer GPUs. It is a step below stable-diffusion.cpp because it runs image models only, with no video generation.

• https://raw.githubusercontent.com/nunchux-ai/nunchaku/main/README.md recorded 2026-09-26

  README: "a high-performance inference engine optimized for 4-bit neural networks"; its news entries name FLUX.1 tools, Kontext and Krea-dev, SANA, Qwen-Image and Qwen-Image-Edit, and Z-Image-Turbo; no video or audio model is named.