stabilityai

/

sdxl-turbo

like 2.27k

Follow

Stability AI 6,011

@joaness I would recommend using something like PonyXL lightning or some finetune of it as they are far more uncensored.
PonyXL is a heavy finetune/retrain of sdxl to make it have a very different prompting style and become very uncensored.
You can use lightning variants so you can generate images very fast, similar to turbo but with higher quality then turbo.

@joaness I used website prebuilt AI that you can add in new model and and loras for things in needs, the model i am using right now is prefect PonyXL model and some lora to make the image for better, i say if the AMD AMUSE Beta is limited so you should find a better more round and better model to make the AI art work better. check out different stable diffussion or stabilityai that is out there or different website that is free or paid. I do believe that YaTharThShaRma999 has some good points as well

@joaness Instructions for modifying an ONNX model used for content filtering in Amuse v2 or greater.

1. Locate and Open the ONNX Modifier Tool:

   • Navigate to C:\Program Files\Amuse\Plugins\ContentFilter in your file explorer.
   • Download the ONNX modifier from Git.
   • Open the ONNX modifier tool.

2. Import the Model:

   • Import the model file named model.onnx located in the same directory (C:\Program Files\Amuse\Plugins\ContentFilter).

3. Modify the Model:

   • Scroll to the bottom of the ONNX modifier interface.
   • Find and click on the “Greater” condition node.
   • Click the “+” next to the B input to add a new value.
   • Enter 1000000 and press Enter.

4. Save Changes:

   • Click the download button in the top left corner to save the modified ONNX model.

These steps will help in adjusting the content filtering settings by modifying the ONNX model directly.