The model I think is fairly self explanatory.

Works well at a strength of ~.7-.8 for more flexibility in posing and such, not entirely sure why poses got heavily baked in when screen captures of the manga had various angles, but it's not too important, I'll investigate that issue for future models later.

Previews made with a goofy comfyui workflow and autismmix confetti. that's different from my normal generic prompt loop after semi-figuring out how to get generation data to embed into the images to make the previews 'functional'. The comfyui workflow should also be embedded into the previews I hope.