Qwen-Image-2512-Fun-Controlnet-Union

Model Card

Name	Description
Qwen-Image-2512-Fun-Controlnet-Union-2602.safetensors	Compared to the previous version of the model, we added Gray control to the model. The model was trained for a longer time than before.
Qwen-Image-2512-Fun-Controlnet-Union.safetensors	ControlNet weights for Qwen-Image-2512. The model supports multiple control conditions such as Canny, HED, Depth, Pose, MLSD and Scribble.

Model Features

• This ControlNet is added on 5 layer blocks. It supports multiple control conditions—including Canny, HED, Depth, Pose, MLSD, Scribble and Gray. It can be used like a standard ControlNet.
• Inpainting mode is also supported.
• When obtaining control images, acquiring them in a multi-resolution manner results in better generalization.
• You can adjust control_context_scale for stronger control and better detail preservation. For better stability, we highly recommend using a detailed prompt. The optimal range for control_context_scale is from 0.70 to 0.95.

Results

Pose + Inpaint	Output

Pose	Output

Pose	Output

Scribble	Output

Canny	Output

HED	Output

Depth	Output

Gray	Output

Inference

Go to the VideoX-Fun repository for more details.

Please clone the VideoX-Fun repository and create the required directories:

# Clone the code
git clone https://github.com/aigc-apps/VideoX-Fun.git

# Enter VideoX-Fun's directory
cd VideoX-Fun

# Create model directories
mkdir -p models/Diffusion_Transformer
mkdir -p models/Personalized_Model

Then download the weights into models/Diffusion_Transformer and models/Personalized_Model.

📦 models/
├── 📂 Diffusion_Transformer/
│   └── 📂 Qwen-Image-2512/
├── 📂 Personalized_Model/
│   └── 📦 Qwen-Image-2512-Fun-Controlnet-Union.safetensors

Then run the file examples/qwenimage_fun/predict_t2i_control.py and examples/qwenimage_fun/predict_i2i_inpaint.py.