My models are not perfect by any means, but with the right workflow and prompting you can really great results.
All models I make in the future will likely be male.
Feel free to message me with any questions or potential requests.
LoRA was trained on Pony. Prompting is just like any other, with the triggers:
"1 boy, Kousei Arima, solo focus, glasses, black hair, blue eyes ".
If you're trying to recreate the exact character from the series, I recommend all of them; otherwise adjust as needed.
Additional optional tags:
"formal attire, necktie"
"casual attire"
I find that AutismMix gives the most consistent results, feel free to use any Pony model and share your results!
Recommended weight is 0.5-0.95, but as the sample pictures show you can adjust as needed.
The images generated were made with the following parameters and a latent upscale:
Cfg: 5.0 - 8.0
Steps: 25
Sampler(s): dpmpp_3m_sde,
Scheduler: Karras
Denoise: 1.00
Lora weight(s): 0.50-0.95
Happy generating!
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
