/vt/ - Virtual Youtubers

>>36737330
That's fine. I use cfg 11 because NAI defaults to it. But the sampler and steps are important as well. As is tag choice.
"city" is shit. "cityscape" is magic.
I took what I think you were trying to do, and tried to do it. Guess which one is city and which one is cityscape?
https://litter.catbox.moe/zesuky.png
https://litter.catbox.moe/7zi4vg.png
I didn't change your prompt very much at all BTW, all the details are below, and it's at the same 512x512 you were using as well:
masterpiece, best quality, robot joints, kazama iroha, hololive, blonde hair, aqua eyes, side ponytail, katana, chest sarashi, city, dutch angle, roboco-san, mechanical arms, mechanical legs, Source Stable Diffusion 1D44365E
{"steps": 28, "sampler": "k_euler", "seed": 1529520539, "strength": 0.69, "noise": 0.667, - that means ENSD 31337 "scale": 11.0,
"uc" that means negative prompt: "lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry"}
>>36737361
I'd honestly not have known any better had you not come back to tell us mate...