Nunchaku is a high-performance inference engine optimized for low-bit diffusion models and LLMs, as introduced in our paper SVDQuant. Check out DeepCompressor for the quantization library.
.. toctree:: :maxdepth: 2 :caption: Installation installation/installation.rst installation/setup_windows.rst
.. toctree::
:maxdepth: 1
:caption: Usage Tutorials
usage/basic_usage.rst
usage/lora.rst
usage/kontext.rst
usage/controlnet.rst
usage/qencoder.rst
usage/offload.rst
usage/attention.rst
usage/fbcache.rst
usage/pulid.rst
.. toctree::
:maxdepth: 1
:caption: Python API Reference
python_api/nunchaku.rst
.. toctree::
:maxdepth: 1
:caption: Useful Tools
:titlesonly:
ComfyUI Support: ComfyUI-nunchaku Plugin <https://github.com/mit-han-lab/ComfyUI-nunchaku>
Customized Model Quantization: DeepCompressor <https://github.com/mit-han-lab/deepcompressor>
Gradio Demos <https://github.com/mit-han-lab/nunchaku/tree/main/app>
.. toctree::
:maxdepth: 1
:caption: Other Resources
faq/faq.rst
developer/contribution_guide.rst