Metadata-Version: 2.1
Name: comfyui-fp8-cublas
Version: 0.0.1
Summary: ComfyUI custom node that patches loaded models (Flux, SDXL, etc.) to run Linear layers in FP8 or MXFP8, unlocking tensor-core throughput on Ada Lovelace (RTX 4090) and Blackwell (RTX 5090) GPUs.
Home-page: https://github.com/marduk191/ComfyUI-fp8_cublas
Requires-Python: >=3.10
Requires-Dist: torchao>=0.10.0
