nunchaku-tech-ComfyUI-nunchaku/docs/source/api/model_patcher.rst at dev - nunchaku-tech-ComfyUI-nunchaku - Gitea: Git with a cup of tea

nodes/nunchaku-tech-ComfyUI-nunchaku

Files

T

Muyang Li a805352a2f feat: support CPU offloading for Qwen-Image to reduce the VRAM usage (#520 )

* update

* offloading implemented; start testing

* check why the results is not correct

* finished the offloading of ComfyUI; start testing it on 5090

* update

2025-09-04 09:52:16 -04:00

8 lines

130 B

ReStructuredText

Raw Permalink Blame History

 model_patcher
 =============
 .. automodule:: comfyui_nunchaku.model_patcher
    :members:
    :undoc-members:
    :show-inheritance: