This Lora was trained on 20 images of the Elon Musk
Prompt: (masterpiece, best quality, awards winner) Elon Musk as muaythai fighter wearing muaythai shorts, epic character composition, by ilya kuvshinov, alessio albi, nina masic, sharp focus, natural lighting, subsurface scattering, f2, 35mm, film grain, <lora:Name:0.85>
Negative promt: cartoon, 3d, ((disfigured)), ((bad art)), ((deformed)), ((poorly drawn)), ((extra limbs)), ((b&w)), weird colors, blurry
Model: Realistic Invision
Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 997014767, Size: 512x704, Model hash: a8e08b582e, Model: realisticVisionV30_v30VAE, Lora hashes: "Name: ca55d948c950"
You can use other prompt and other models it gives interesting and unpredictable results
This Lora in video:
Description
FAQ
Comments (3)
The face for some reason came out with a lot of contrast
I mean...I think base models know how to draw Elon Musk well enough, probably can get better results without the LoRA unless you deliberately added those artefacts. Trained on 20 images, 1 epoch, 20000 steps, you meant you were using 1000 repeats???
https://civitai.com/articles/680/characters-clothing-poses-among-other-things-a-guide
I hope you will see this comment and these articles can help you.