Hey there,
this time I have another cyberpunk art LoRA for you. I tried to combine the hightech, scifi visuals of cyberpunk with the impressionistic art style.
The LoRA is trained on cyberpunk themed images in orange and teal coloring. So it's has a natural bias towards this look, but with proper prompting you should be able to easily change the coloring or the theme of the image.
If you enjoy my work, consider showing your support with a š or ā¤ļø on the model or imagesāit really keeps me motivated!
You can also follow me or buy me a coffee ā at: https://ko-fi.com/madcaddie
Thanks and have fun!
Description
Usage tips:
It inherited the trigger 'mad-cybrpnkimprss' from the SDXL version, but it's not really necessary, since it works basically the same without.
I recommend a strength of 1.
FAQ
Comments (3)
please explain in more detail what is the difference between pony and sdxl version?
Firstly I'm not an expert on neutral networks and therefore might use the wrong vocabulary.
Pony and SDXL are two different base models, that both are based on the SDXL structure.
Pony uses different style prompts and prompting in general works somewhat differently. Please check out the model page.
That means you could use the SDXL LoRA for Pony (and vice versa), but it will likely not produce the same results. So since Pony is quite popular I have decided to train specific Pony LoRAs.
Lastly, the final effect for a LoRa, even if you use the same training data, will be different for each base model.
I hope that answers your question
@maDcaDDieĀ this looks interesting, I have never seen anything like this before, itās worth trying to feel the effect