Alibaba Cloud PAI Releases Z-Image-Turbo-Fun-Controlnet-Union - Multi-Condition ControlNet Model

news

Alibaba Cloud PAI team releases Z-Image-Turbo-Fun-Controlnet-Union, a ControlNet model supporting multiple control conditions including Canny, HED, Depth, Pose, and MLSD, trained at 1328 resolution

Alibaba Cloud PAI team has recently released Z-Image-Turbo-Fun-Controlnet-Union, a ControlNet model designed for Z-Image-Turbo.

Model Information

The model supports the following control conditions:

  • Canny - Edge detection
  • HED - Holistically-nested edge detection
  • Depth - Depth map
  • Pose - Human pose
  • MLSD - Line segment detection

The model was trained on 1 million high-quality images at 1328 resolution for 10,000 steps.

ComfyUI Support

ComfyUI is currently working on supporting this model. Users can follow ComfyUI updates and will be able to use it in workflows once official support is available.

Usage

The model can currently be used through the VideoX-Fun repository. The development team plans to train on more data and add inpaint mode support in the future

Alibaba Cloud PAI Releases Z-Image-Turbo-Fun-Controlnet-Union - Multi-Condition ControlNet Model | ComfyUI Wiki