Billy's Cardboard Cars - SDXL LoRA
This time our attempt was to train a more toy-like cardboard cars. Happens that then we decided to try motorcycles, and we expected the model to be able to produce other vehicles, and IT DOES! Like airplanes, spaceships, ships... and even trains!
Model Description:
This LoRA model adapts Stable Diffusion XL (SDXL) to generate images of vehicles constructed from cardboard. While the trigger word is Cardboard Cars, the model is capable of generating a wide variety of ground vehicles - including trucks, motorbikes and more - in the cardboard aesthetic. Experimentation may reveal it also produces other vehicle types, such as airplanes and ships.
Trigger Word: Cardboard Cars
Recommended LoRA Weight: 0.8
Model Type: LoRA
Base Model: Stable Diffusion XL (SDXL)
License: CC-BY-4.0 (Creative Commons Attribution 4.0 International) - This license allows for commercial and non-commercial use, modification, and distribution, provided attribution is given to the creator (Robb-0).
How to Use:
1. Load the SDXL base model into your preferred Stable Diffusion environment.
2. Apply this LoRA model.
3. Use the trigger word "made of cardboard" (for general purposes) or "cardboard cars" (for ground vehicles) in your prompt.
4. Experiment with different prompts and LoRA weights (starting with 0.8) to achieve desired results.
5. Consider adding detail to your prompt. For example, “a cardboard car driving down a busy street, photorealistic”
Example Prompts:
* Cardboard Cars, a red sports car, sunny day
* Cardboard Cars, a dump truck full of toys, detailed, realistic
* Cardboard Cars, a futuristic motorbike, cyberpunk, neon lights
* a spaceship made of cardboard, detailed, intricate
Known Limitations:
* The model may produce unexpected vehicle types beyond cars.
* Results may vary depending on the complexity of the prompt and the base model settings.
* May require prompting with modifiers to specify need details and styles.
Disclaimer:
This model is provided “as is” without any warranty. The creator is not responsible for any misuse or unintended consequences of using this model.