Trained on Realistic Vision, wanted to experiment this concept. I have to say this model is not perfect, the flexibility is not very high (my dataset is not diverse enough, couldn't find many high-res bubble popping images). But if you are patient with your prompts and parameters, you will get interesting results.
I have gotten some decent results with txt2img, but it is still better to use with Controlnet or Inpainting.
Suggested Weight: 1.0 - 1.6, depending on your prompting style
I would be very happy to see what you can do with this model.😃
If you like this model, please consider leaving a like. Or you can buy me my next meal here. Thanks.❤️