comfy-aimdo 0.2.14
misc/py-comfy-aimdo
ComfyUI: AI Model Dynamic Offloader for PyTorch
Description
AI Model Dynamic Offloader (AIMDO) is a PyTorch VRAM allocator that implements on-demand offloading of model weights when the primary PyTorch VRAM allocator comes under pressure. This enables running larger models on GPUs with limited VRAM by intelligently managing memory and offloading weights to system RAM when needed.
Dependencies
- build devel/py-build
- build devel/py-installer
- build devel/py-setuptools
- build devel/py-setuptools-scm
- build devel/py-wheel
- build lang/python311
- run lang/python311
- run misc/py-pytorch
Commit History
may be incomplete — full history at freebsd-ports on GitHub
Loading commit history — this may take up to a minute on first view. Reload the page in a moment.