Training on "red skin".
I have been working to integrate various unique features into LoRa's (and now LoCon). I have found making individual trainings on each feature, then merging the ones needed together gets remarkable results.
This has been trained on about 30 AI generated images from various models (picked out from hundreds), after using a LoRa made on about 1700 images of characters with red skin. This helped to eliminate the style bias and focus on just the feature (in this case red skin).
This can be used stand-alone to add red skin with much higher efficiency, or mixed with other LoRA / LoCon to mix the feature in. I recommend to avoid using this with multiple other LoRA / LoCon without mixing first, it is more prone to cause issues.
If you like this, get me a coffee :D