Just a random fella doing random things in his spare time. Feel free to use the model or download it.Any feedback is welcome!It was an interesting experiment on using a very limited and repetitive dataset.The lora was trained using a trigger word:<ul><li>silpheed</li></ul>Other useful prompts:<ul><li>purple skin, red eyes, green hair, spiked hair</li><li>long sleeves, forehead accessory</li></ul>The model is a bit rigid as a result of the limited dataset. I would advise to use a strenght of 0.7-0.8 as standard to avoid odd winds effect etc.Common Notes:Trained on base Illustrious.It should work with other Loras without strange conflicts.It should work on NAI too (mostly on the epsilon pred). If it gives some anatomy aberration, try without the main trigger word

Silpheed [Yu-Gi-Oh!]

Trained on anima_baseV10Strength between 0.6-1.0 usually does not give troubles. It is better using between 0.7-0.8 as it is a bit overfitted due to the restricted dataset.It should work with other Loras without strange conflicts.

Files with SHA256: ee8148eb6a0919b8193a71c70095521278341be17b0b90f6944ea5a307717c16