Controlnet Python Packages

reshot

ReShot — copy the shot, not the actors. Video → depth map, OpenPose skeleton or canny lines for Seedance 2.0/2.5 reference video, MiniMax H3 Fun ControlNet and Wan VACE. Apache-2.0. Open-sourced by Maosika 猫斯卡 (www.maosika.com), the AI short-drama production system.

ppdiffusers

Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks, including end-to-end large-scale multi-modal pretrain models and diffusion model toolbox. Equipped with high performance and flexibility.

aisketcher

Model-independent sketch-to-image studies: seed scouting, controlled variation, replayable manifests, and a zero-download local tour.

armature-studio

Image-to-video with a GLB instead of an image. Block the shot in headless Blender - character, pose, camera - render per-frame control sequences, and the video model paints life over it: film, cutscenes, character performance, any footage, with one persistent main character.

poseorbit

Find the people in a picture, pick whose pose, turn it in 3D, frame it, and draw the skeleton your image model follows. CPU only, no torch.

comfydl

ComfyDL is a robust Command Line Interface (CLI) tool for downloading models for ComfyUI. It automates the process of fetching models from various sources (like Hugging Face, Civitai) and placing them into the correct directories within your ComfyUI installation.

omniinfer-client

Python SDK for Stable Diffusion API (Txt2Img/Img2Img/ControlNet/VAE)

diffusersplus

This project is under development.

custom-diffusion

This project is under development.

landmarkdiff

Facial surgery prediction from a single photo today, 3D reconstruction from phone video tomorrow. Diffusion-based, open-source, clinically aware. MICCAI 2026.