Aoi Futaba | Cat Planet Cuties

详情

下载文件 (1)

模型描述

Aoi Futaba from Cat Planet Cuties

Trained with 200+ images from the anime and some fanart on the PonyDiffusionV6 checkpoint. The samples were generated with AutismMix.

Recommended settings:

  • Sampler: DPM++ 2M Karras / Euler a - 25-35 steps

  • CFG: 7

  • Resolution: 1024*1024, 832*1280, 1280*1024, (works with most common XL resolutions)

  • Lora weight: 0.8

  • HighRes: 4x_NMKD-Siax_200k, upscale by 1.5x, 25 steps, 0.35 denoising

Basic prompt:

<lora:AoiFutaba:0.8> aoifutaba, 1girl, long hair, black hair, green eyes, glasses, shorts

Costumes:

EP1 Battlesuit

Prompt: ep1battlesuit, aoifutaba,1girl, bodysuit, helmet, small breast

  • Add "spear" or "holding spear" to the prompt for her weapon

  • If you want to see her face trough the helmet try to generate "upper body" or "close-up" shots

  • If you want to generate images with just the suit: Remove "helmet" from your prompt and describe her features (black hair, green eyes, ... etc)

EP12 Spacesuit

Prompt: ep12spacesuit, aoifutaba,1girl, bodysuit, white gloves, white boots, small breasts

  • Didn't have many good shots of this suit so the outputs are a bit hit or miss.

  • I recommend putting "megaman" and "neon genesis evangelion" in the negative prompt.

Military uniform / Operator

Prompt: aoifutaba, military uniform, 1girl, smile, short hair, black hair, gloves, green eyes, glasses, fingerless gloves, load bearing vest

  • Yup, stable diffusion still can't draw proper guns...

Bunny suit

Prompt: aoifutaba, 1girl, bunny ears, yellow beanie, micro bikini, flat chest, long hair, black hair, green eyes

Additional notes:

  • Adding "source_anime" and "anime screencap" to your prompt will give your images that neat flat look, but it's not mandatory.

  • Make sure to describe the background/setting in detail, otherwise you'll get plain images.

  • Don't forget about the Pony's usual quality tags.

  • My samples have the prompts embedded if you need inspiration. :D

此模型生成的图像