Resized to 25% of original (view original)
Prompt | {{{moriya suwako}}}, 1girl, solo, cowboy shot, mamemochi, sitting, wariza, {arms behind back}, dynamic angle, smile, {{{{toddler}}}}, {{{{{{oppai loli}}}}}}, {huge breasts}, [medium hair], blonde hair, yellow eyes, [parted bangs], white bikini, white thighhighs, {yellow hat}, {pyonta}, red hair ribbon, {bedroom}, bed |
---|---|
Negative prompt | monochrome, adult, long body, button, {neck ribbon}, bad hands, missing fingers |
Sampler | k_euler |
Seed | 3445606647 |
Steps | 20 |
Cfg Scale | 5 |
Model Hash | c1e1de52 |
Moriya Suwako
It's been quite a while since I did image generation with NovelAI.
I did it because I heard that it had evolved quite a bit a while ago, and I was surprised to find that it was able to generate much better than I expected.
This is an order of magnitude more effective, especially in terms of prompting.
It is fascinating to be able to do so much without LoRA or even modifications.