Model Settings

• Steps: 8

• CFG: 1.0

• Samper: euler

• Scheduler: sgm_uniform

The 4steps version feels closer to the Turbo Version. The 8steps version is closer to the original model.

What to expect

This model can do good images of cities, buildings, and photorealistic people. It lost some qualities of the base SinteticoXL like text rendering, or good logos.

Sometimes the model may output written text when prompted for it, but it's far from perfect. It's actually random luck to get decent text.

It is still best to use the base model with the Lightning LoRA.

Please, read this article on the current state of Lightning merges.

Why I made this

I wanted to build a really fast version of my model. This one can generate 9 images (1024x1024) in approximate 1 second on RTX 3070 TI.

The quality of this model is close to the base one on version 1.2. Please refer to the base SinteticoXL for better quality.

How can you use

Complex prompting is not required, but can be helpful to guide towards your vision.

The base model is already merged with FreeU and Self-Attention Guidance. Adding those will break the results.

Commercial usage of generated images is allowed.

You are solely responsible for any content that you create using this model. In addition, your use of this model implies that you accept an agreement to not use it to produce harassing, harmful, illegal, or otherwise highly-objectionable imagery.

You may not resell this model or provide it as a service.