all of these were finetuned based off of kani-anime as a base model. namori based was trained off of danbooru images by the artist namori. yryr test was trained off of anime screenshots on top of namori base. yuruyuri prototype was trained off of additional yryr art official and non official and was trained on top of yryr test. these were all trained with 768 images and batch size 16 1.6-e5 learning rate in the stable tuner trainer.
comparison of all the models
prompted with the prompt
masterpiece, best quality, 1girl, by namori, yuru yuri, toshinou kyouko, blonde hair, red hair bow, smile, blue eyes, nanamori school uniform 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, artist name, bath, onsen, water Steps: 18, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 666, Size: 768x1024, Model hash: 3b339a4d, Denoising strength: 0.7, First pass size: 384x512
"by namori" "yuru yuri" should help shift it more towards this style you can also prompt almost all of the characters on the models with yryrtest or yuruyuriprototype.