# SVDQuant nunchaku runtime (FLUX / Qwen quantized inference)
#
# Do NOT install the unrelated PyPI package named "nunchaku" (data segmentation).
# Use the NodeTool registry wheels or install from GitHub releases.
#
# Pick the wheel matching your Python, torch, and CUDA versions:
# https://nodetool-ai.github.io/nodetool-registry/simple/nunchaku/
#
# Example (Python 3.11, torch 2.9, CUDA 12.8, Windows):
# pip install "https://github.com/nunchaku-tech/nunchaku/releases/download/v1.2.1/nunchaku-1.2.1+cu12.8torch2.9-cp311-cp311-win_amd64.whl"
