How to Use midjourney-v4-diffusion Model on OnnxDiffusersUI

A new Stable Diffusion model fine tuned on Midjourney v4 images has been published recently. It’s easy to use it on OnnxDiffusersUI, Follow these simple steps:

1. Open up a command prompt and activate the virtual environment. On my computer, the commands are

   cd OnnxDiffusersUI
   .\virtualenv\Scripts\activate.bat
   

2. Download and convert the model. Note that this step takes a while.

   python convert_stable_diffusion_checkpoint_to_onnx.py --model_path="prompthero/midjourney-v4-diffusion" --output_path="model/midjourney_v4_diffusion_onnx"
   

3. Run onnxUI.py to use the model

   python onnxUI.py --model="model/midjourney_v4_diffusion_onnx"
   

That’s it. Remember to add “mdjrny-v4 style” in your prompt to use the style.

Here are some examples:

Stable Diffusion 1.5

midjourney-v4-diffusion

Stable Diffusion 1.5

midjourney-v4-diffusion