Overview
ControlNet adds extra conditioning to diffusion models so image generation can follow visual signals such as Canny edges, depth maps, normal maps, line drawings, and pose-related inputs. The official repository includes pretrained model workflows, annotation scripts, Gradio examples, and instructions for training custom ControlNets. It is a developer and researcher tool commonly used with Stable Diffusion workflows. The main implementation is published under Apache-2.0. The repository is not archived, but its latest meaningful push is older than 12 months, so this directory marks the project unmaintained while the code remains available.